METHOD FOR CALIBRATING A WHEEL SENSOR AND CORRESPONDING WHEEL SENSOR AND RAILWAY TECHNICAL SYSTEM WITH SUCH A WHEEL SENSOR

Description

[0001] In the field of railway automation, wheel sensors are used, for example, for wheel detection and axle counting in track vacancy detection systems. Wheel sensors are also used for other switching and signaling tasks, such as train-induced switching on and off of level crossing safety systems or train announcements. Corresponding wheel sensors, which can be based on different operating principles, typically detect rail vehicle wheels based on their iron mass. Depending on the sensor type, it is necessary to calibrate the wheel sensor to define the influence that the wheel sensor is already subjected to by the rail in the absence of a wheel to be detected as the base state.In practice, this process must be repeated at regular intervals, for example, to compensate for wear on the rail or to take this into account by calibrating the wheel sensor. Furthermore, other changes and influencing factors may also require recalibration of a wheel sensor.

[0002] Calibrating wheel sensors typically requires work on the track. Maintenance personnel ensure that the wheel sensor to be calibrated is unaffected and perform the calibration or calibration required for the respective sensor type. This is not a major effort during the initial installation of the wheel sensor, as the personnel are already on-site. However, the situation is different for recalibrations that are required regularly or cyclically, as the presence of maintenance personnel is required solely for the calibration of the wheel sensor.

[0003] European patent EP 2 289 757 B1 discloses a method for calibrating a wheel sensor of a track vacancy detection system. In this method, a control computer of the track vacancy detection system transmits a calibration release signal to the wheel sensor upon the track vacancy detection section monitored by the wheel sensor becoming clear. Upon receipt of the calibration release signal, the wheel sensor determines whether calibration is required, and if so, performs it. This known method enables automatic calibration of wheel sensors even without the presence of maintenance personnel in the track area.The calibration release signal ensures that calibration of the respective wheel sensor only takes place when there is a sufficiently long time window in which the calibration can be carried out without endangering ongoing train operations.

[0004] Document CA 2 685 575 A1 discloses a rail-mounted train detection system and describes a sensor detection device and method based on the principle of diversity. The intelligent wheel flange sensor system uses sensors on both rails, separated by an optimal distance to accommodate an entire bogie. This enables the detection of rolling stock, maintenance vehicles, and locomotives based on the different wheel sequences as the vehicle moves over the sensors. Time, space, and wheel diversity are captured when measuring each axle. Train progress is determined by integrating the speed-time curve.A self-test function dynamically tests the sensors through programmable magnetic self-test stimuli that emulate a full train movement at any speed across the sensors, so that all system responses to this stimulus are identical to those of a real train.

[0005] The present invention is based on the object of providing a particularly flexible method for automatically calibrating or measuring a wheel sensor.

[0006] This object is achieved according to the invention by a method for calibrating a wheel sensor having the features of claim 1, wherein the wheel sensor determines that a calibration is to be carried out, the wheel sensor determines a time suitable for carrying out the calibration, and the wheel sensor carries out its own calibration at the determined time.

[0007] According to the first step of the inventive method for calibrating a wheel sensor, the wheel sensor determines that calibration is required. This means that the wheel sensor itself checks, based on criteria known to it, whether calibration is necessary at a given time or not. The relevant decision is thus made decentrally by the respective wheel sensor itself.

[0008] According to the second step of the method according to the invention, the wheel sensor determines a suitable time for performing the calibration. This means that the wheel sensor not only independently decides whether calibration is necessary, but also automatically determines the appropriate time for performing the calibration. The wheel sensor determines the appropriate time for calibration without the involvement of other components. This means, in particular, that the wheel sensor does not receive any messages or signals from other components, such as a higher-level control computer, in this context.

[0009] According to the third step of the method according to the invention, the wheel sensor calibrates itself at the determined time. The actual calibration, like the preceding method steps, is performed automatically, i.e., without requiring intervention by maintenance personnel.

[0010] The method according to the invention thus enables automatic calibration of the wheel sensor, with all method steps being carried out by the wheel sensor itself, so that it calibrates completely automatically. This offers the particular advantage that additional, higher-level components or communication with such components are not required to perform the calibration of the wheel sensor. The method according to the invention can therefore be used flexibly, regardless of the respective function and integration of the respective wheel sensor into the respective railway system. In particular, the method according to the invention does not require the wheel sensor to receive data, messages, or signals from another component.Thus, the method according to the invention is particularly suitable for calibrating wheel sensors that only have a unidirectional interface for transmitting counting pulses or travel messages, for example to an axle counting computer, but cannot receive signals.

[0011] It should be noted that, within the scope of the present invention, the terms "calibration" and "calibration" refer to any activities and processes in which the wheel sensor is measured or adjusted relative to an unaffected state. This involves a new basic setting of the wheel sensor that goes beyond simple drift tracking. For example, the calibration of the wheel sensor can include setting or readjusting switching thresholds or other operating parameters of the wheel sensor relative to one or more measured variables determined in the unaffected state of the wheel sensor.

[0012] Furthermore, it should be noted that the first and second steps of the method according to the invention can also be swapped in time or can run completely or partially in parallel. This means that the wheel sensor can first determine a suitable time for carrying out the calibration and only then can it be checked whether calibration actually needs to be carried out. Furthermore, the corresponding test can, for example, also comprise a plurality of sub-steps, at least one of which can be carried out before and at least one after the determination of the suitable time for carrying out the calibration. Furthermore, it is also possible for the determination of the suitable time for carrying out the calibration to comprise a plurality of sub-steps, at least one of which is carried out before and at least one after the check as to whether calibration needs to be carried out.

[0013] To avoid any misunderstandings, it should be noted at this point that, within the scope of the present invention, any device for detecting wheels or axles of a rail vehicle is referred to as a wheel sensor. This applies regardless of the respective operating principle and regardless of whether the device in question comprises only one component arranged on one side of a rail or several components arranged on one or both sides of a rail. In the latter case, the wheel sensor can, for example, have at least one transmitting device and at least one receiving device. Depending on the respective embodiment and use, wheel sensors within the meaning of the present invention are also known by the terms "counting point," "axle counting point," or "rail contact."

[0014] The method according to the invention is designed such that the time suitable for performing the calibration is determined by the wheel sensor based on the detected end of a rail vehicle's travel past the wheel sensor. This is advantageous because after the end of a travel past the wheel sensor, e.g., after a train has completed its passage, a sufficiently long time window is generally available for calibration, during which no further travel is expected, taking into account the minimum train headway.

[0015] According to the invention, the wheel sensor detects the end of the rail vehicle's travel over the wheel sensor, taking into account a temporal sequence of detections of the rail vehicle's wheels. When a rail vehicle, such as a train, passes over, several wheels are detected by the respective wheel sensor within a specific period of time. Taking into account the temporal sequence of detections of the rail vehicle's wheels, the wheel sensor can determine the end of the travel over the wheel sensor and thus determine that a suitable time has arrived for performing calibration.

[0016] Advantageously, the method according to the invention can be further developed such that the end of the rail vehicle's travel past the wheel sensor is detected by the wheel sensor if no further wheel is detected by the wheel sensor within a predetermined time period or a time period that can be determined by the wheel sensor. Advantageously, the predetermined time period or a time period that can be determined by the wheel sensor is selected such that it is significantly longer than the pauses between the detection of individual wheels of the rail vehicle, so that it can be concluded with high reliability that the rail vehicle has actually completely passed the wheel sensor.

[0017] According to another particularly preferred embodiment of the method according to the invention, the speed of the rail vehicle and / or a change in speed of the rail vehicle is determined by the wheel sensor and taken into account when detecting the end of the travel past the wheel sensor. This is intended, in particular, to prevent the wheel sensor from incorrectly assuming that the rail vehicle has already completely passed the wheel sensor due to the rail vehicle decelerating or coming to a standstill. The speed or change in speed of the rail vehicle can be determined by the wheel sensor, for example, based on the time interval between the detection of the rail vehicle's wheels.Increasing time intervals indicate a deceleration of the rail vehicle, which, depending on the time between the detections of the individual wheels, may be an indication that the rail vehicle is coming to a standstill or has come to a standstill in the area of ​​the wheel sensor. Depending on the type of wheel sensor in question, the speed of the rail vehicle can be determined from the length of the wheel pulses detected by the wheel sensor or from the rise and / or fall speed of the edges of the wheel pulses. It should be noted that an exact speed determination is not necessary, since the only aim is to detect situations in which the rail vehicle is traveling very slowly or has come to a standstill in the area of ​​the wheel sensor.In this case, after detecting the corresponding situation, the wheel sensor can advantageously postpone a due calibration by identifying the respective time as unsuitable for performing the calibration. The same can also apply, for example, if a resting value recorded by the wheel sensor differs before and after a test run, as this could be an indication that a rail vehicle wheel may be resting on the wheel sensor or in the wheel sensor's area.

[0018] The wheel sensor can determine whether calibration is necessary in various ways. Within the scope of the method according to the invention, it is only important that the wheel sensor determines the corresponding calibration requirement completely autonomously.

[0019] Advantageously, the method according to the invention can also be configured such that the wheel sensor determines that calibration is required based on a comparison of at least one measured value with at least one target value. This is advantageous because a corresponding comparison can ensure needs-based calibration. This means, in particular, that calibration of the wheel sensor is only performed when it is actually necessary due to a deviation in the at least one measured value, which preferably refers to the unaffected state of the wheel sensor.

[0020] According to a further particularly preferred development of the method according to the invention, a temporal average of the at least one measured value is calculated, and the temporal average of the at least one measured value is compared with the at least one target value. Using a temporal average of the at least one measured value ensures that individual measured values ​​deviating from the target value do not immediately result in a calibration of the wheel sensor.

[0021] Preferably, the temporal average of the at least one measured value is calculated over a period of one day or several days. This is advantageous in that, in this case, time-of-day dependent effects, such as temperature dependence, are also averaged out, preventing corresponding effects from triggering calibrations of the wheel sensor. By averaging the at least one measured value over one day, i.e., a period of exactly or at least substantially 24 hours, or several days, i.e., substantially a multiple of 24 hours, corresponding time-of-day dependent effects are advantageously taken into account in a simple manner.

[0022] According to a further particularly preferred embodiment of the method according to the invention, the wheel sensor determines that calibration is to be performed if the comparison of the at least one measured value with the at least one target value results in a deviation within a predetermined value range. This means, on the one hand, that the deviation must exceed a certain minimum value in order to trigger calibration of the wheel sensor. This is based on the consideration that even the smallest deviations between the at least one measured value and the at least one target value, which may, for example, be a measured value recorded during a previous calibration, should not lead to calibration of the wheel sensor.Furthermore, depending on the circumstances, calibration of the wheel sensor may also be omitted if the deviation between the at least one measured value and the at least one target value is too large. In this case, the deviation exceeds the "calibratable range," making calibration of the wheel sensor inappropriate. Reasons for this could include, for example, a malfunction of the wheel sensor, interference with the wheel sensor by a wheel, or the wheel sensor falling off the rail.

[0023] Alternatively or in addition to the previously mentioned preferred developments of the method according to the invention, this can advantageously also be designed such that the wheel sensor determines that a calibration is to be carried out if a predetermined period of time has elapsed since the last calibration. A corresponding temporal condition can be used on its own or linked to a comparison of at least one measured value with at least one target value, so that a calibration of the wheel sensor is only carried out or required, for example, if a predetermined period of time has elapsed since the last calibration and, at the same time, a deviation of a measured value of the wheel sensor from a target value is detected.

[0024] The method according to the invention is applicable to both single-channel and dual-channel wheel sensors. Dual-channel wheel sensors, also referred to as dual sensors, are typically used for directional detection of rail vehicles.

[0025] Advantageously, the method according to the invention can be designed such that, in a wheel sensor with two sensor channels, the method is carried out independently in each of the two sensor channels. This offers the advantage that each of the two sensor channels can independently decide on the need and timing of its calibration. This avoids, in particular, dependencies between the sensor channels that could potentially compromise the reliability of the wheel sensor.

[0026] As an alternative to the aforementioned embodiment, the method according to the invention can advantageously also be further developed such that, in the case of a wheel sensor with two sensor channels, the method is carried out across both sensor channels. This offers the advantage that, on the one hand, coordination between the two sensor channels is possible so that they do not perform a calibration simultaneously. This can be advantageous, for example, in that the other sensor channel can detect unforeseen traffic even during ongoing calibration, which can be advantageous with regard to the signaling reliability of the wheel sensor.In addition, it is also possible in principle to combine information from both sensor channels during the calibration of the wheel sensor, for example to determine the speed of a passing rail vehicle and to take this into account when deciding when a suitable time for a required calibration is present.

[0027] The invention further relates to a wheel sensor, in particular for a track vacancy detection system according to claim 11. With regard to the wheel sensor, the object of the present invention is to provide a wheel sensor which supports a particularly flexible method for automatically calibrating or measuring the wheel sensor.

[0028] This object is achieved according to the invention by a wheel sensor, in particular for a track vacancy detection system, wherein the wheel sensor is designed such that it determines that a calibration is to be carried out, determines a time suitable for carrying out the calibration and carries out the calibration of itself at the determined time.

[0029] The advantages of the wheel sensor according to the invention correspond to those of the method according to the invention, so that reference is made to the corresponding above statements in this regard.

[0030] According to a particularly preferred development, the wheel sensor according to the invention is designed to carry out the method according to one of the previously described preferred developments of the method according to the invention. The advantages of the respective development of the wheel sensor according to the invention correspond to those of the corresponding preferred development of the method according to the invention, so that reference is also made to the corresponding above explanations in this regard.

[0031] The invention further comprises a railway technology according to claim 13.

[0032] The invention will be explained in more detail below using exemplary embodiments. Figure 1 shows a schematic sketch of signals detected by a wheel sensor during travel by a rail vehicle to explain an embodiment of the method according to the invention, and Figure 2 shows a flow chart to further explain the embodiment of the method according to the invention.

[0033] Figure 1To explain an exemplary embodiment of the method according to the invention, a schematic diagram shows signals detected by a wheel sensor during a passage by a rail vehicle. Signals S 1 to S 32 are shown, which are detected by a wheel sensor during a passage by a rail vehicle. The term "passage" is to be understood as meaning that the wheels of the rail vehicle move through a detection zone of the wheel sensor and are consequently detected by it. The corresponding wheel sensor can be used for various switching and control tasks within the framework of automated railway operations and, in particular, can be a component of a track vacancy detection system. According to the illustration in the Figure 1the signals S 1 to S 32 are detected by the wheel sensor at times t 1 to t 32 . In the context of the described exemplary embodiment, it is assumed, by way of example, that the rail vehicle consists of eight vehicles, each with four axles, whereby two axles can be combined in a bogie. In this case, for example, the signals S 1 to S 4 , S 5 to S 8 , S 9 to S 12 , S 13 to S 16 , S 17 to S 20 , S 21 to S 24 , S 25 to S 28 and S 29 to S 32 could be traced back to wheels of a corresponding vehicle or vehicle part of the rail vehicle, for example in the form of a locomotive or a carriage.

[0034] In Figure 1It can be seen that the maximum time interval between the successive signals S 1 to S 32 , which is designated by Δt max, occurs between the detection of the second and the third axle of the first vehicle. The maximum axle distance within a rail vehicle is usually around 20 m, so that the maximum interval Δt max between the successive signals S 1 to S 32 can be deduced from the respective speed of the rail vehicle or an expected speed thereof. It should be noted that the maximum time interval Δt max could of course also occur for another vehicle in the rail vehicle, depending on a possible change in speed of the rail vehicle and the respective vehicle types.

[0035] If the wheel sensor has determined that its own calibration is required, it can determine a suitable time for performing the calibration. The wheel sensor can derive the time at which a calibration should be initiated or is to be initiated essentially from the "occupied" / "free" status sequences of one or, in the case of a dual-channel dual sensor, both sensor systems. In the latter case, the decision regarding the time at which a calibration is permissible or suitable for calibration can be made separately in each subsystem or channel of the wheel sensor or axle counting point, or from the results of both subsystems. It is important that other components, such as an axle counting computer in the internal system, are not involved in determining the suitable time for performing the calibration.This is particularly advantageous in cases where it avoids changing the interface between the respective wheel sensor and the internal system.

[0036] In the embodiment of the Figure 1after the last signal S 32 at time t 32 up to a time t 33 no further travel, i.e. no further axle, is detected by the wheel sensor. The pause Δt e after the last axle of the rail vehicle has been detected by the signal S 32 is therefore significantly longer than the maximum time interval Δt max between the successive signals S 1 to S 32 . The wheel sensor uses this as a criterion to detect the end of travel by the rail vehicle in question over the wheel sensor and thus to determine time t 33 as a suitable time for carrying out the calibration. Due to normal train headways this ensures that no travel over the wheel sensor is to be expected during the calibration and therefore the calibration process does not impair the detection and reporting of a passage by a following rail vehicle.

[0037] Regardless of this, the wheel sensor is advantageously designed in such a way that it can correctly report wheel crossings even during calibration, or at least detect them and issue an error message. Different configurations are possible depending on the specific wheel sensor design. For example, even the shortest wheel crossings, i.e., at the highest speed encountered, typically take about 3 ms. As long as the wheel sensor calibrates more quickly or its detection capability is not impaired by calibration, performing the calibration has no effect on the wheel sensor's reliability.If a situation arises in which the wheel sensor fails to perform an error-free count due to or during its calibration, this does not lead to a safety-critical situation in the case of a single error, at least in a wheel sensor of an axle counting system, even if, contrary to expectations, the wheel sensor is driven over during calibration. While this could result in an axle counting error, potentially resulting in a disruption to train operations, such an axle counting error is not safety-critical, provided it can be reliably ruled out that the wheel sensor is "blind" to a complete pass of a rail vehicle.

[0038] According to the above explanations, the time suitable for carrying out the calibration can be determined by the wheel sensor based on a detected end of a passage of the wheel sensor by a rail vehicle and then the wheel sensor can carry out its own calibration at the determined time. As can be seen from the representation of the Figure 1As explained, the end of the rail vehicle's travel over the wheel sensor can be detected by the wheel sensor, taking into account a temporal sequence of the detection of wheels of the rail vehicle. Specifically, the end of the rail vehicle's travel over the wheel sensor can be detected by the wheel sensor, for example, when no further wheel is detected by the wheel sensor within a predetermined time period Δt e that can be determined by the wheel sensor. The time period can be predetermined in such a way that it is set to a constant value, for example, 45 s.

[0039] Alternatively, the time period can also be determined, for example, by the wheel sensor itself. In this case, it is conceivable, for example, that the time period Δt e is a multiple of the maximum distance Δt max detected by the wheel sensor between the successive signals S 1 to S 32. For example, the time period Δt e can be selected as five or ten times the maximum time interval Δt max or can be determined by the wheel sensor itself. If necessary, the braking capacity of the trains traveling on the respective route can also be taken into account in order to be able to reliably rule out, even in the event of a braking of the respective rail vehicle in the area of ​​the wheel sensor in question, that the rail vehicle has braked sharply in the area of ​​the wheel sensor and possibly come to a standstill, thus preventing the wheel sensor from passing through.If necessary, the speed of the rail vehicle and / or a change in its speed can also be determined by the wheel sensor and taken into account when detecting the end of the travel of the wheel sensor. After the connection with... Figure 1 In particular, it has been explained how the wheel sensor can determine a suitable time for carrying out the calibration. Figure 2 the embodiment of the method according to the invention will be explained in more detail.

[0040] Figure 2 shows a flow chart to further explain the embodiment of the method according to the invention.

[0041] In the context of the algorithm described below as an example, it is assumed that in a method step 10 it is first checked whether a predetermined period of time has passed since the last calibration. The predetermined period of time can be, for example, one month or three months, in which case the calibration of the wheel sensor would be carried out at most monthly or at most every three months. If the condition is met, i.e. the predetermined period of time has passed since the last calibration, the method continues with step 20; otherwise it returns to the starting point, for example by checking at regular intervals whether the predetermined period of time has passed since the last calibration.

[0042] In method step 20, the wheel sensor checks whether calibration is necessary, taking into account a comparison of at least one measured value with at least one target value. In the described exemplary embodiment, it is assumed that in a method step 30, the wheel sensor calculates a temporal average of a measured value in the form of a standard or rest voltage of the wheel sensor, and that this average is fed to the test step 20 as an input variable in a method step 40. Advantageously, a temporal average is calculated over a period of one or more days in order to average out, in particular, daytime-dependent effects, such as temperature effects.

[0043] In method step 20, the wheel sensor checks whether a deviation within a predetermined value range exists by comparing at least one measured value in the form of the mean value of the rectified voltage in the idle state with a target value, for example in the form of the rectified voltage value determined during the last calibration. This check ensures that the deviation in question is neither too large nor too small, and that the wheel sensor is therefore within a "calibratable range." If this is the case, the method proceeds to method step 50. Otherwise, the method returns to the starting point, since calibrating the wheel sensor taking the measured value into account is neither necessary nor appropriate.

[0044] In method step 50, after determining in the previous steps that calibration is required, the wheel sensor determines a suitable time for performing the calibration. This means that the wheel sensor checks whether it is occupied by a wheel of a rail vehicle or whether such occupancy or travel is expected. According to the explanations in connection with the Figure 1This can be achieved, for example, by the wheel sensor detecting the end of a rail vehicle passing over the wheel sensor and determining the respective time as a suitable time for calibration. If the wheel sensor has determined the respective time as suitable for its own calibration, the wheel sensor performs its own calibration in method step 60. Otherwise, the wheel sensor checks again, particularly the next time it is passed over by a rail vehicle, whether or when a suitable time for calibrating the wheel sensor exists.

[0045] After the wheel sensor has been calibrated in method step 60, a time counter is reset or a timestamp is set in method step 70. This information can then be used by the wheel sensor the next time it runs through method step 10 to decide whether calibration of the wheel sensor is necessary from a time perspective.

[0046] According to the above statements in connection with the embodiment of the method according to the invention, this and a wheel sensor designed to carry out the method have the particular advantage that the calibration of the wheel sensor is fully controlled and carried out by the wheel sensor itself. This means, in particular, that the determination of a suitable time for calibration is also carried out solely by the wheel sensor. The corresponding automatic, self-calibration of the wheel sensor has the advantage, on the one hand, that, since the need for calibration is determined by the wheel sensor itself, predetermined fixed inspection intervals can be eliminated. Furthermore, the independent execution of the calibration by the wheel sensor can advantageously also lead to an increase in the availability of the wheel sensor and thus of the railway system of which it is a component.Furthermore, since no information transfer from other components to the wheel sensor is required to carry out the calibration, the method can also be advantageously applied to situations or wheel sensors in which only a unidirectional interface between the wheel sensor and, for example, an axle counting computer is available, so that a corresponding information transfer would not be possible or would require a change in the interface, for example to the internal system.

Claims

1. Method for calibrating a wheel sensor, wherein - the wheel sensor establishes that a calibration procedure is to be performed (10, 20), - the wheel sensor ascertains (50) a point in time that is suitable for the calibration procedure to be performed and - the wheel sensor performs (60) the self-calibration at the ascertained point in time, characterised in that the point in time that is suitable for the calibration procedure to be performed is ascertained (50) by the wheel sensor with reference to the fact that it has been identified that a rail-borne vehicle has completely driven past the wheel sensor by taking into consideration a temporal sequence of detecting wheels of the rail-borne vehicle.

2. Method according to claim 1, characterised in that the wheel sensor identifies that a rail-borne vehicle has completely driven past the wheel sensor insofar as the wheel sensor does not detect (50) a further wheel within a predetermined time period or within a time period that may be determined by the wheel sensor.

3. Method according to claim 2, characterised in that the wheel sensor ascertains the velocity of the rail-borne vehicle and / or a change in the velocity of the rail-borne vehicle and takes this into consideration (50) when identifying that the rail-borne vehicle has completely driven past the wheel sensor.

4. Method according to one of the preceding claims characterised in that by taking into consideration a comparison of at least one measured value with at least one desired value, the wheel sensor establishes that a calibration procedure is to be performed (20).

5. Method according to claim 4, characterised in that a temporal mean value of the at least one measured value is formed (30, 40) and the temporal mean value of the at least one measured value is compared (50) with the at least one desired value.

6. Method according to claim 5, characterised in that the temporal mean value of the at least one measured value is formed over a time period of one day or of multiple days (30, 40) .

7. Method according to one of claims 4 to 6, characterised in that the wheel sensor establishes that a calibration procedure is to be performed insofar as the comparison of at least one measured value with the at least one desired value results (20) in a deviation that lies in a predetermined value range.

8. Method according to one of the preceding claims, characterised in that the wheel sensor establishes that a calibration procedure is to be performed insofar as a predetermined time period has elapsed since the last calibration procedure (10).

9. Method according to one of the preceding claims, characterised in that in the case of a wheel sensor having two sensor channels the method is performed in each of the two sensor channels independently of one another.

10. Method according to one of claims 1 to 8, characterised in that in the case of a wheel sensor having two sensor channels the method is performed across both sensor channels.

11. Wheel sensor, in particular for a track clear signalling installation, wherein the wheel sensor is configured in such a manner that - it establishes that a calibration procedure is to be performed (10, 20), - it ascertains (50) a point in time that is suitable for the calibration procedure to be performed and - it performs (60) the self-calibration procedure at the ascertained point in time, characterised in that the point in time that is suitable for the calibration procedure to be performed is ascertained (50) by the wheel sensor with reference to the fact that it has been identified that a rail-borne vehicle has completely driven past the wheel sensor by taking into consideration a temporal sequence of detecting wheels of the rail-borne vehicle.

12. Wheel sensor according to claim 11, characterised in that the wheel sensor is configured so as to perform the method according to one of claims 2 to 10.

13. Railway installation, in particular a track clear signalling installation, having at least one wheel sensor according to one of claims 11 or 12.