EP3789264B1 - Method and device for slip detection and rail vehicle - Google Patents

Description

The invention relates to a method for determining a correction value while simultaneously excluding slip events on one or more axles of a ground-based vehicle. The invention further relates to a device for carrying out such a method and to a ground-based vehicle with such a device.

Rail vehicles are ground-based vehicles for which a vehicle speed and a path derived from it must be determined precisely and safely in manual or automatic driving mode. Speed sensors or path incremental encoders are generally used for this purpose, the latter of which are assigned to a wheel or wheelset of the rail vehicle. However, such sensors cannot directly detect a wheel skidding or sliding, so that the wheel can rotate faster or slower than the rail vehicle is actually traveling. As a result, a speed measured by the sensors may not correspond to the current vehicle speed of the rail vehicle. The term "correction factor" is also used below as a synonym for the correction value.

From the AU 2005234619 A1 is a method for controlling the speed of a rail vehicle and from the DE 10 2005 010 118 A1 A control device of a rail vehicle is known.

An object of the present invention is to provide a method for slip detection in a wheelset for a rail vehicle, which enables a reliable and precise determination of a vehicle speed of a rail vehicle and which can help to keep delays in the operation of the rail vehicle to a minimum.

The task is solved by a method for slip detection with simultaneous calculation of correction values of different axles for a ground-based vehicle, here a rail vehicle according to the independent patent claim. Advantageous Embodiments of the method are specified in the dependent patent claims.

According to one aspect of the invention, a method for slip detection in a wheelset for a rail vehicle, to which a first wheelset with a first speed sensor and a second wheelset with a second speed sensor are assigned, comprises detecting a plurality of measurement signals from the first and second speed sensors, each of which is representative of a speed of the first and second wheelset of the rail vehicle. The method further comprises determining a respective speed value for the first and second wheelset, each as a function of a respective measurement signal from the first and second speed sensors, respectively. The method further comprises forming speed value pairs, each with a determined speed value for the first and second wheelset. The method further comprises comparing the respective speed values for the first and second wheelset of a formed speed value pair with one another and determining a uniform correction factor for all determined speed value pairs as a function of the comparison. The method further comprises determining speed deviations based on the uniformly valid correction factor and determining slip of the first and/or second wheel set depending on the speed deviations. The method also comprises, in multiple applications, determining a plurality of correction factor deviations depending on the determined correction factors and determining slip of the first and/or second wheel set based on the determined correction factor deviations.

The method described enables reliable detection of sliding or skidding of a wheel set and can contribute to an accurate determination of the vehicle speed. The method enables a robust relative scale determination or correction factor determination with reliable exclusion of sliding and skidding. Such a correction factor determination takes into account the determined correction factors of different track sections and the resulting speed deviations. In this way, slip areas in the form of outliers can be reliably detected based on the speed deviations and correction factors calculated in the process. The correction factors and correction factor deviations enable the determination of speed deviations of a wheelset in order to obtain equality in relation to the speed of the wheelsets when using the correction factor.

According to a preferred development, the method comprises controlling a vehicle speed of the rail vehicle depending on the determined slip of the first and/or second wheel set. Controlling the vehicle speed of the rail vehicle can in particular comprise providing a safety margin for a permissible vehicle speed of the rail vehicle depending on the determined slip, so that controlling the vehicle speed of the rail vehicle is also carried out depending on the provided safety margin.

• Ungenaue Messung des Raddurchmessers
• Abnutzung während der Fahrt
• Änderung des Reifeninnendrucks, zum Beispiel bei Gummibereifung
• Beladung, vornehmlich bei Gummibereifung.

For ground-based vehicles, such as rail vehicles, it is necessary to determine the vehicle speed and the resulting travel path precisely and reliably in both manual and automatic driving modes. This can be achieved, for example, by using position incremental encoders that are assigned to a respective wheel or wheel set of a rail vehicle. It is a finding in connection with the present invention that even with such sensors, sliding or skidding of a wheel set is not always reliably detected. The wheel of the wheel set can, for example, rotate faster than the rail vehicle is actually traveling (skidding) or it can rotate slower than the rail vehicle (sliding). With position incremental encoders, the distance travelled or the speed of the wheel and the rail vehicle is usually determined using the diameter and the circumference of the wheel derived from it. If the diameter of a wheel changes, the wheel and vehicle speed calculated from the wheel revolutions changes. Causes for incorrect or inaccurate wheel diameter determination include:

• Inaccurate measurement of the wheel diameter
• wear and tear during the journey
• Change in tire pressure, for example in rubber tires
• loading, especially with rubber tires.

If a speed determination changes, for example due to an incorrect or inaccurate wheel diameter, speed differences are caused between different position incremental encoders. Such differences can be incorrectly interpreted as sliding or skidding and treated as alleged slip. If slip is assumed to be present, specified safety margins for speed intervals are increased because it is more difficult to estimate the actual vehicle speed. This causes a false safety-related reduction in the vehicle speed, so that the rail vehicle now travels more slowly and considerable delays can be caused.

For a reliable determination of the vehicle speed and safe operation of the rail vehicle, it is therefore advantageous to reliably detect any slip (sliding or skidding) of the wheel or wheelset and to incorporate this into the control of the vehicle speed. Using the method described, reliable slip detection can be carried out so that delays of a rail vehicle can be counteracted and a planned and punctual operation of the rail vehicle.

According to a further development, the method comprises discarding one of the determined correction factors depending on the determined speed deviations per correction factor or correction factor deviation between correction factors, and controlling a vehicle speed of the rail vehicle depending on the remaining correction factors. The determined speed deviations of the calculated correction factors and the correction factor deviations make it possible to precisely identify the phases and time ranges in which a wheel slip occurs.

For example, there is a speed difference between the two wheel sets that would justify a correction factor of around 1.1. If a correction factor of, for example, 4 is determined for a short time, this correction factor deviates significantly from the others and can be identified as slip. For detection, a uniform correction factor, for example 2.55, is determined and gaps or deviations in the individual correction factors per difference are shifted into respective speed deviations. The further the deviation from the uniformly calculated correction value, the greater the speed deviations. The correction factors calculated one after the other for route sections and/or the speed deviations accordingly show a clear peak, which enables reliable slip detection. If the correction factors described were now averaged, this would cause an unjustified adjustment of the vehicle speed. A safety margin would incorrectly be applied to the vehicle speed.

Using the described method, such outliers caused by slippage can be reliably detected and correctly evaluated. For example, such a If the value of the correction factor is discarded, it is no longer included in the evaluation and the vehicle speed of the rail vehicle and a safety margin can be specifically adjusted.

According to the invention, determining the correction factor involves a regression analysis. Statistical analysis methods can be used to achieve the best possible averaging of the determined correction factors, for example by minimizing the resulting speed deviations. Such a regression analysis and compensation calculation can be based on the least squares method, for example.

According to a further development of the method or one of the described further developments of the method, the method comprises specifying a deviation threshold value for a tolerable speed deviation and comparing the determined speed deviations with the specified deviation threshold value. The method accordingly also comprises determining a slip of the first and/or second wheel set depending on the comparison of the determined speed deviations with the specified deviation threshold value. In particular, a slip range can thus be recognized on the basis of a tolerable speed deviation in the form of a deviation threshold value, which should not be exceeded in accordance with operation.

According to a further development of the method or one of the described further developments of the method, the first and the second speed sensor each have a position incremental encoder. Accordingly, determining a respective speed value for the first and the second wheel set comprises providing circumference values for a wheel of the first wheel set and a wheel of the second wheel set, to which one of the position incremental encoders is assigned to each. The method further comprises detecting a number of revolutions of the respective wheel of the first wheel set and the second wheel set and determining a respective speed value for the first wheel set and for the second wheel set depending on the respective circumference values provided and the respective detected number of revolutions.

According to a further development of the method or one of the described further developments of the method, determining a slip of the first and/or second wheel set as a function of the determined speed deviations includes recognizing slip-influenced and slip-uninfluenced areas of the first and/or second wheel set as a function of the determined correction factors and/or the determined speed deviations. The control of a vehicle speed of the rail vehicle then also takes place as a function of the recognized slip-influenced and slip-uninfluenced areas of the first and/or second wheel set.

According to a further development of the method or one of the described further developments of the method, the method includes forming an arithmetic mean of the determined correction factors. The control of a vehicle speed of the rail vehicle is therefore also carried out depending on the arithmetic mean of the correction factors formed. In particular, the detected slip areas are not included in the formation of the arithmetic mean, so that a falsification of the averaging is counteracted. Alternatively, another averaging method or another value formation method can be used for a summary correction factor, so that, for example, a weighting of the correction factors is also taken into account.

The arithmetic mean of the correction factors can also be stored for a subsequent travel cycle of the rail vehicle. Controlling a vehicle speed The rail vehicle is then controlled, for example, depending on the stored arithmetic mean of the correction factors and is loaded, for example, when operations start and is directly included in the control of the vehicle speed.

The described method and the described further developments of the method enable reliable and precise slip detection at a finely granular level. No additional sensors are required and no generous safety margins are required, which are included in advance in the speed and path calculation due to the possible occurrence of slip in order to counteract corresponding sliding or skidding processes using control technology. This means that no general safety margins are to be assumed and it can contribute to increased accuracy in determining the vehicle speed of the rail vehicle.

For example, the method requires only two position incremental encoders that are mounted on non-mechanically coupled axles of the rail vehicle. The vehicle speed and the travel path for the rail vehicle are determined using just two position incremental encoders. Data on the wheel circumferences, for example, are provided. However, since the wheel circumferences are only known to a limited extent, the raw data recorded using the measurement signals from the position incremental encoders for the calculated speeds of the first and second wheel sets deviate from the actual vehicle speeds of the rail vehicle. These speed deviations can be accepted up to a specified size.

However, these deviations may be too large for reliable slip detection and are determined and then eliminated. This enables particularly precise and reliable slip detection. The detection of slip makes it possible in particular to areas not affected by slip. In areas not affected by slip, general safety margins in relation to the vehicle speed for a possible occurrence of slip, which could not be determined conventionally or not with sufficient accuracy, can be significantly reduced. Rail vehicles such as trains can then run on time according to the timetable or at least undesirable delays in operations can be reduced.

The described method and the further developments enable a high degree of usability regardless of the type of rail vehicle and the route it travels. The method implements a generic approach and does not require any information about the route, such as longitudinal gradient. No information about a current acceleration or braking process is required either.

The described method and the further training also ensure a high level of acceptance among users, as no interaction on the part of the user is necessary. The method can be carried out in an automated manner. In addition, a performance gain for timetable compliance can be achieved due to reduced delays and higher line utilization. In addition, the method provides a clear approach that can contribute to maintenance-friendly operation of a rail vehicle.

In addition, the method is cost-effective. Existing position incremental encoders can be used to carry out the method and the method can be combined with other approaches to detect systematic deviations, such as sliding and slipping processes in position incremental encoders. Considering the large number of rail vehicles used, significant cost savings can be achieved. In addition, the method is relatively future-proof, as there are few to no returns. are to be expected. Returns include customer complaints and manufacturer improvements.

The method described contributes in particular to improved accuracy in terms of skidding and skidding detection and meets high accuracy requirements in terms of determining vehicle speed. It enables the evaluation of highly redundant information through adjustment and statistical tests, among other things with the aim of calculating a correction factor with high accuracy and then automatically detecting systematic errors in order to be able to exclude them so that they are not included in the final correction factor determination. Furthermore, the method described enables a flexible accuracy target with regard to the scale and thus also the speed.

A further aspect of the invention relates to a device for detecting slip in a wheel set for a rail vehicle, which is designed to carry out one of the methods described above. The device is implemented, for example, as an on-board computer or control unit of the rail vehicle, which has a computing unit and a data storage unit and enables evaluation and processing of the recorded measurement signals and the determined speed values. Alternatively or additionally, such a device is implemented as a backend or external server unit that can communicate with the rail vehicle using signals.

According to a further aspect of the invention, a rail vehicle comprises a first wheel set and a second wheel set which are mechanically decoupled from each other. The wheel sets are coupled via one or more car bodies of the rail vehicle, but there is no direct mechanical coupling between the wheel sets, for example by means of a transmission. The rail vehicle further comprises a first speed sensor and a second speed sensor which are connected to a are assigned to the respective wheel set. The speed sensors each have a position incremental encoder and the measurement signals from the speed sensors are each representative of a speed of the first wheel set and the second wheel set. The rail vehicle also has the device described above, which is coupled to the speed sensors in terms of signal technology.

Because the device and the rail vehicle are designed to carry out one of the previously described methods, described properties and features of the method are also disclosed for the device and for the rail vehicle and vice versa.

Figur 1

eine schematische Darstellung eines Schienenfahrzeugs mit einer Vorrichtung zur Schlupferkennung für mindestens ein Radsatz des Schienenfahrzeugs,

Figur 2

ein Ablaufdiagramm für ein Verfahren zur Schlupferkennung für mindestens ein Radsatz des Schienenfahrzeugs,

Figur 3

ein weiteres Ablaufdiagramm für ein Verfahren zur Schlupferkennung für mindestens ein Radsatz des Schienenfahrzeugs, und

Figur 4

beispielhafte algorithmische Zusammenhänge zum Verarbeiten bei der Durchführung des Verfahrens zur Schlupferkennung.

The above-mentioned properties, features and advantages of the invention and the manner in which they are achieved are explained in more detail by the following description of the embodiments of the invention in conjunction with the corresponding figures. They show:

Figure 1

a schematic representation of a rail vehicle with a device for slip detection for at least one wheel set of the rail vehicle,

Figure 2

a flow chart for a method for slip detection for at least one wheel set of the rail vehicle,

Figure 3

another flow chart for a method for slip detection for at least one wheel set of the rail vehicle, and

Figure 4

exemplary algorithmic relationships for processing when carrying out the slip detection procedure.

Figure 1 shows a schematic side view of a rail vehicle 1 with a control unit 3, which is coupled in terms of signal technology to a first speed sensor 5 and a second speed sensor 6. The speed sensors 5 and 6 are position incremental encoders for the wheel sets 7 and 8. The wheel sets 7 and 8 have respective wheels and axles that are not mechanically directly coupled to one another. The speed sensors 5 and 6 enable the detection of measurement signals that are each representative of a speed of the first wheel set 7 and the second wheel set 8 of the rail vehicle 1. As explained below, the position incremental encoders mounted on the axles can be used to advantage to enable reliable and precise detection of sliding and skidding of the wheels of the first and/or second wheel sets 7 and 8.

A method for slip detection in one of the wheel sets 7, 8 for the rail vehicle 1 can be carried out in accordance with Figure 2 or the one in Figure 3 shown flow chart in the control unit 3. In a step S1, a plurality of measurement signals of the first speed sensor 5 and the second speed sensor 6 are detected.

In a step S2, a respective speed value for the first wheel set 7 is determined depending on a respective measurement signal from the first speed sensor 5 and a respective speed value for the second wheel set 8 is determined depending on a respective measurement signal from the second speed sensor 6. Furthermore, speed value pairs are formed, each with a determined speed value for the first and second wheel sets 7, 8. If a predetermined number of usable speed value pairs have been determined or formed, the method is continued in a step S3. For example, every half second a measurement signal from the first and second speed sensors 5 and 6 is recorded for a period of one minute, so that 120 speed value pairs are formed.

In step S3, the respective speed values for the first and second wheel sets 7, 8 of a respective speed value pair are compared with each other and a correction factor is determined for all speed value pairs depending on the comparison.

In a step S4, speed deviations are determined depending on the determined correction factor. In particular, a statistical model can be used for the evaluation. In particular, in step S4, a regression analysis or a compensation calculation is carried out, which is based, for example, on the method of least squares, which carries out a minimization method with regard to the smallest possible deviation of the determined speed deviations with regard to a correction factor for all pairs of values.

For example, the following mathematical relationships can be formulated for each determined speed pair: v_measured_Sensor1_i = correction factor * v_measured_Sensor2_i. The mathematical value "correction factor" only exists once and is automatically set by the minimization method. For "v_measured_Sensor1_i", deviations "v_measured_Sensor1_deviation_i" of the type v_measured_Sensor1_i + v_measured_Sensor1_deviation_i = correction factor * v_measured_Sensor2_i are determined in such a way that the sum of the squares of these deviations is minimal.

In step S5, a slip of the first and/or second wheel set 7, 8 is determined as a function of the determined deviations for the measured speeds with respect to a calculated correction factor and the vehicle speed of the rail vehicle 1 is controlled as a function of the detected slip, taking into account the previously calculated correction factor.

The slip detection procedure can also be carried out according to the flow chart illustrated in Figure 3, which shows the variable relationships according to Figure 4 In a step S1, it is checked whether the default conditions of the variables to be processed are met. If this is not the case, the process continues in a step S2. If the default conditions are met, the process continues in step S3.

In step S3, measurement signals from the position incremental encoders are recorded and speed values for the respective axles or wheel sets 7 and 8 are determined. Furthermore, speed value pairs are formed, each containing a speed value for the first and second wheel sets. If a predetermined number of evaluable speed value pairs have been formed, the method is continued in step S5. Otherwise, the method is terminated or started again in step S1.

In step S5, a k _{PSSA section} factor is determined or adjusted. The k _{PSSA section} factor represents a correction factor that is calculated for all pairs of measured values, for example 120 pairs of measured values that form a section, by minimizing the speed deviations with respect to a sensor. The following relationships apply in particular: v_measured_Sensor1_i + v_measured_Sensor1_deviation_i = correction factor * v_measured_Sensor2_i, where "i" runs from 1 to 120, for example. The correction factor k _{PSSA section} factor is estimated or calculated by the adjustment. The acronym "PSSA" refers to the English term "pseudo static scale adjustment" and can be referred to in German as pseudo-static scaling adjustment.

In a step S7, the mathematical and stochastic model underlying the adjustment is checked. The mathematical model includes the assumption that there is only one scale factor k _PSSA-Section for, for example, all 120 velocity pairs and the stochastic model includes the Assumption regarding the accuracy of the measured speeds. Furthermore, the accuracy of the correction factor calculated from the previous step S5 is compared with a predetermined threshold value. If a desired accuracy is met and the assumptions in the mathematical and stochastic model are not refuted, the process can be continued in a step S9. If a desired accuracy is not met and/or the assumptions of the model tests are refuted, the process is continued in a step S8.

In step S8, the current data and determined speed values are discarded. The process is terminated and restarted when new speed values arrive in step S1.

In step S9, it is checked whether kfactorsInSeries are available. For example, kfactorsInSeries has the value 3 here. If valid correction factors have been calculated 3 times in a row, the process continues in step S11. Otherwise, the correction factor k _PSSA-Section is saved, the process is terminated and then restarted in step S1 when new measured values arrive.

The variables kfaktorsInSeries represent a number of valid correction factors calculated one after the other. A correction factor is calculated from, for example, 120 pairs of values. It can happen that a valid correction factor was calculated in the first minute and none in the following minute. Then there is a valid correction factor. However, 3 are needed to continue the process. In the third minute, for example, a valid one is calculated. If 3 are available, the process continues.

In step S11, all possible combinations of two of the determined k _{PSSA section} factors are set up. With three k _{PSSA section} factors, there are 3 combinations of two: 12, 13 and 23.

With a value of 4 for the kfactorsInSeries, the combinations would be: 12, 13, 14, 23, 24 and 34.

In a step S13, it is checked whether the calculated correction factors of the respective combinations are the same in the sense of agreeing within tolerable deviations. This includes comparing the determined correction factors. If the correction factors agree within a specified tolerance range, the process is continued in a step S15. Otherwise, the process is terminated and started again in step S1. Each combination comprises two different correction factors that are compared with each other. With a value of 3 for the kfactorsInSeries, there are 3 independently calculated correction factors that can have been created within 3 minutes or extend over a longer period of time, for example 148 minutes.

In step S15, an arithmetic mean is calculated and the k _PSSA factor is provided for a real-time method for skidding and spin exclusion. The k _PSSA factor is the arithmetic mean of the k _{PSSA section} factors . If the value for the kfactorsInSeries is 3, there are exactly 3 of them.

The described method relates to a quasi-static scaling adjustment and is preferably used in conjunction with a method for sliding and spin exclusion. The control unit 3 implements a processing unit that enables a scale of the position incremental encoder to a referencing position incremental encoder. When carrying out the described method, in particular the Figure 4 The algorithmic relationships between different variables shown in the slip detection are taken into account as follows:
The first, topmost graph according to Figure 4 shows the vehicle speed of the rail vehicle 1 over time according to the measurement signals of the first and second position incremental encoder. One of the two position incremental encoders functions as a reference incremental encoder.

The second graph according to Figure 4 shows a corresponding temporal representation of the calculated k _{PSSA section} factors. The representation indicates that only speed value pairs are considered for further processing which have a value that is greater than a predetermined threshold value v _{PSSA limit} . The v _{PSSA limit} is a threshold value. In particular, measured value pairs are only formed if both measured speeds are above this value. One reason for such a useful procedure is of an error-technical nature. A correction factor can be calculated with high accuracy if the vehicle is traveling fast enough. If the 120 measured value pairs of a section were also filled with low speeds, the target accuracy for k _PSSA factor in step S7 might not be achieved.

The third graph shows whether the k _PSSA factors of the combinations are equal (point below the upper line) or unequal (point above the upper line). Each point represents a combination. The lower lines or lines in the lower area of the third graph show whether the k _PSSA factor is generally valid, see S7, to be used for combination formation. If there is no line in the area, this means that the corresponding factor should not be processed further. The lower line or line is derived from the other graphs, the fourth and fifth graphs, which Figure 4 If there are lines below the solid line, a corresponding lower line is drawn in the third graph.

The fourth graph shows the statistically determined deviation according to a statistical test based on the results of the least squares method (χ ² test). A test for each area is fulfilled if the specified limit or the specified deviation threshold (visualized by a solid line) is not exceeded by the determined values (illustrated as individual line sections).

In the fifth graph of Figure 5, counted from the top, an accuracy of the k _{PSSA section} factors is shown for each area. A corresponding requirement is met if the accuracy of the k _{PSSA section} factor (individual line sections) does not exceed a specified accuracy limit for the k _{PSSA section} factor according to the variable σ _{kPSSAsection limit} shown (solid line).

The sixth graph shows that the k _PSSA-Section factor deviates significantly from zero. The k _PSSA-Section factor corresponds to the correction factor described above.

In particular, the relationships shown in the third graph are relevant for the method described. The line sections shown below the points show the range of k _{PSSA section} factors that satisfy the mathematical stochastic model analysis and a specified accuracy. These k _{PSSA section} factors are used to compare the correction factors. Outliers are discarded and not taken into account.

If the k _PSSA-Section factors in the possible combinations formed from factors calculated from kfactorsInSeries are statistically equal, the points corresponding to the k _PSSA-Section factors are arranged in the third graph below the upper solid line. The areas used for this comparison are the line sections shown below the points that are assigned to the k _{factorsInSeries} .

The k _{PSSA section} factors are the correction factors; correction factor deviations are generally only calculated for non-linear problems in order to iteratively arrive at the correct correction factor. In this case, however, it is a linear problem, see equation, where the correction factor is calculated directly. This calculation creates deviations in the measured speeds of a sensor because, for example, there can only be one correction factor for 120 pairs of values. These deviations are used to determine whether slippage (sliding/skidding) occurs and the section is then discarded (steps S7 and S8). For example, it is necessary to ask how much speed should be added to or subtracted from the first sensor, for example, so that the speed of the reference sensor is multiplied by the correction factor. One optimization criterion is the minimization of such deviations squared. However, other criteria are also acceptable for an optimization calculation, such as absolute values.

Although the invention has been illustrated and described in detail using embodiments, the invention is not limited to the disclosed embodiments and the specific combinations of features explained therein. Further variations of the invention can be obtained by a person skilled in the art without departing from the scope of the patent claims.

list of reference symbols

1

rail vehicle

3

control unit of the rail vehicle

5

first speed sensor / position incremental encoder

6

second speed sensor / position incremental encoder

7

first axle / first wheelset of the rail vehicle

8

second axle / second wheelset of the rail vehicle

S(i)

Step of a method for slip detection and/or correction factor determination in a wheelset for a rail vehicle

t

Time

v

vehicle speed

Claims (11)

1. Method for slip detection in a wheelset (7, 8) for a rail vehicle (1), to which a first wheelset (7) with a first speed sensor (5) and a second wheelset (8) with a second speed sensor (6) are assigned, comprising:

   - capturing a plurality of measurement signals of the first speed sensor (5) and the second speed sensor (6), which are in each case representative of a speed of the first wheelset (7) and the second wheelset (8) of the rail vehicle (1),

   - ascertaining a respective speed value for the first wheelset (7) as a function of a respective measurement signal of the first speed sensor (5) and for the second wheelset (8) as a function of a respective measurement signal of the second speed sensor (6),

   - forming a specified number of speed value pairs with an ascertained speed value for the first and the second wheelset (7, 8) in each case,

   - comparing the respective speed values for the first and the second wheelset (7, 8) of each formed speed value pair with one another and ascertaining a unified correction factor over the specified number of formed speed value pairs by way of a regression analysis as a function of the comparison,

   - ascertaining a speed deviation for one of the wheelsets (7, 8) as a function of the ascertained unified correction factor by way of a comparison of the speed value of the one wheelset (7, 8) with the speed value of the other wheelset (7, 8) multiplied by the ascertained unified correction factor, and

   - ascertaining a slip of the first and/or second wheelset (7, 8) as a function of the ascertained speed deviation.
2. Method according to claim 1, comprising:
   controlling a vehicle speed of the rail vehicle (1) as a function of the ascertained slip of the first and/or second wheelset (7, 8) while taking into consideration the correction factor.
3. Method according to claim 2, in which the controlling of the vehicle speed of the rail vehicle (1) comprises:

   - providing a safety uplift for a permitted vehicle speed of the rail vehicle (1) as a function of the ascertained slip, and

   - controlling the vehicle speed of the rail vehicle (1) as a function of the provided safety uplift.
4. Method according to one of claims 1 to 3, comprising:

   - ascertaining at least one further unified correction factor for a further specified number of speed value pairs as a function of the comparison of the respective speed value pairs for the first and the second wheelset (7, 8) of the formed further speed value pairs with one another,

   - discarding one of the ascertained unified correction factors as a function of the speed deviations ascertained in each case here, and

   - controlling a vehicle speed of the rail vehicle (1) as a function of the remaining unified correction factor.
5. Method according to one of claims 1 to 4, comprising:

   - specifying a deviation threshold value for a tolerable speed deviation,

   - comparing the ascertained speed deviations with the specified deviation threshold value, and

   - ascertaining a slip of the first and/or second wheelset (7, 8) as a function of the comparison of the ascertained speed deviations with the specified deviation threshold value.
6. Method according to one of claims 1 to 5, in which the first and the second speed sensor (5, 6) in each case have a path incremental encoder and the ascertaining of a respective speed value for the first wheelset (7) and for the second wheelset (8) comprises:

   - providing circumference values for a wheel of the first wheelset (7) and a wheel of the second wheelset (8), to which a path incremental encoder is assigned in each case,

   - capturing a number of revolutions of the respective wheel of the first wheelset (7) and the second wheelset (8), and

   - ascertaining a respective speed value for the first wheelset (7) and for the second wheelset (8) as a function of the circumference values provided in each case and the number of revolutions captured in each case.
7. Method according to one of claims 1 to 6, in which the ascertaining of a slip of the first and/or second wheelset (7, 8) as a function of the ascertained speed deviations comprises:
   detecting slip-influenced and slip-uninfluenced regions of the first and/or second wheelset (7, 8) as a function of the ascertained speed deviations, and

   - controlling a vehicle speed of the rail vehicle (1) as a function of the detected slip-influenced and slip-uninfluenced regions of the first and/or second wheelset (7, 8).
8. Method according to one of claims 1 to 7, comprising:

   - ascertaining at least one further unified correction factor for a further specified number of speed value pairs as a function of the comparison of the respective speed values for the first and the second wheelset (7, 8) of the formed further speed value pairs with one another,

   - forming an arithmetic mean of the ascertained unified correction factors, and

   - controlling a vehicle speed of the rail vehicle (1) as a function of the formed arithmetic mean of the ascertained unified correction factors.
9. Method according to claim 8, comprising:

   - storing the arithmetic mean of the correction factors for a subsequent travel cycle of the rail vehicle (1), and

   - controlling a vehicle speed of the rail vehicle (1) as a function of the stored arithmetic mean of the correction factors.
10. Apparatus (3) for slip detection in a wheelset (7, 8) for a rail vehicle (1), which is configured to perform a method according to one of claims 1 to 9.
11. Rail vehicle (1), having:

    - a first wheelset (7) and a second wheelset (8), which are mechanically decoupled from one another,

    - a first speed sensor (5) and a second speed sensor (6), which are assigned to a respective wheelset (7, 8) and each have a path incremental encoder, and the measurement signals thereof in each case are representative of a speed of the first wheelset (7) and the second wheelset (8), and

    - an apparatus (3) according to claim 10, which is coupled to the speed sensors (5, 6) from a signal perspective.

Images