DE19953677C1 - Rail vehicle derailment detection method, uses detection of acceleration of rail vehicle component in vertical or transverse direction relative to rail track - Google Patents

Abstract

The derailment detection method detects the acceleration of a rail vehicle component in direct or indirect contact with the rail track (14), in a direction which is vertical or transverse to the track direction (12), for comparison of its double integral with an upper and/or lower limit value, for indicating the derailment of the rail vehicle (10). An Independent claim for a rail vehicle derailment detection device is also included.

Description

The invention relates to a method and a direction for detecting derailment of a track bound vehicle.

It is known that track-bound vehicles, at for example, passenger cars, freight cars, motor vehicles witness, subways, S-Bahn, trams or the same by means of wheel sets along a lane, especially of rails, are movable. The wheel sentences roll over and over the rails on the one hand take a lead of the vehicles and on the other hand, a transfer of a drive relationship approximately braking energy. It is known that as a result of technical problems, such as material fatigue or the like, or from on the Rail obstacles, there is an ent track of the vehicles can come. This is the rule moderately with high material and where appropriate personal damage associated. In particular, it can Derailing a single wheel set of a train for  Cause a chain reaction, as a result derail several vehicles of the train and elevate cause damage.

Acceleration sensors are generally known. This are used in various areas of technology uses one on assemblies or the like to detect acting acceleration. The Be acceleration sensors usually include schwin the seismic masses that are suspended in a loading deflect the effect of acceleration. The deflection of the seismic mass is recorded and from this one of the acting acceleration proportional output signal provided.

From DE 25 07 645 A1 an arrangement for Acquisition, transmission and monitoring of measured values known, by means of the derailed axes of a track bound vehicle can be recognized. For this is proposed on a axle bearing one Railway car to provide a measuring device that one Accelerometer, by means of which the Derailment of an axis occurring strong shock should be recognized. This one from the Accelerations resulting from vibrations if a stored limit value is exceeded checked. In the known arrangement it is disadvantageous that an evaluation of the accelerations is only strong can be done incorrectly since no output conditions exist. So it is early Detection of derailment not possible.

The invention has for its object a Ver drive and a device of the generic type specify by means of which a Ent track of a vehicle early on by a ver better evaluation of measured acceleration signals can be recognized.

According to the invention, this object is achieved by a ver drive ge with the features mentioned in claim 1 solves. By accelerating one with the Lane directly or indirectly in contact Component of the track-bound vehicle vertically and / or is determined transversely to a direction of travel and the acceleration with an acceleration sensor is measured and one supplied by it Signal is integrated twice over time and this double integrated acceleration signal an upper and / or lower limit (Gu, Gs) is compared and when falling below and / or Exceeding the limit on a derailment is recognized is advantageously possible immediately after a derailment occurs, it is sure to detect. This takes advantage of the fact that if a wheel set is derailed, this - from Ab come from the lane - an acceleration experiences vertical and / or transverse to the direction of travel, the as the occurrence of an event, here an Ent track, is evaluable. By the double Integration of the acceleration signal can Evaluation of - also changing - defined Measurement conditions take place from.  

According to the invention, the object is further achieved by a Device with the noted in claim 12 paint solved. The fact that one directly or indirectly component in contact with the lane track-bound vehicle an acceleration sensor is assigned, the acceleration sensor with a Data acquisition and evaluation unit is connected, the data acquisition and evaluation unit Signal delivered by the acceleration sensor twice integrated over time, and the processed signal with at least one upper and / or lower compares the limit value and if it falls below and / or exceeding the limit an action, in particular braking the track-bound driving stuff that can be triggered can be easily which during the intended use of the track-bound vehicles on which the acceleration component equipped with the inclination sensor, in particular vertically and transversely to a direction of travel accelerations, detect and if necessary, a corresponding countermeasure Initiate derailment.  

He further preferred embodiments of the invention give up from the rest, in the subclaims mentioned features.

The invention is in one embodiment example with reference to the accompanying drawings explained. Show it:

Fig. 1 is a schematic side view of a chassis of a railroad car;

Fig. 2 shows a schematic curve of a doubly integrated acceleration signal;

Fig. 3 shows schematically in a diagram the determination of the length of integration intervals and

Fig. 4 schematic characteristic curves for determination of the initial parameters of an integration interval.

Fig. 1 shows schematically a vehicle 10 which is movable on a track 14 in the direction of travel 12 . The vehicle 10 is, for example, a freight car or passenger car of a train, a locomotive or the like of a train, a tram car, subway car, S-Bahn car or the like. The track 14 is known to consist of two rails arranged parallel to each other, which are arranged on a corresponding substructure. The track 14 forms a lane 16 for the vehicle 10 . The vehicle 10 comprises at least one bogie 18 which is connected to a body 22 of the vehicle 10 via a secondary suspension system 20 . According to further exemplary embodiments, the vehicles do not have a secondary suspension system 20 . The bogie 18 is assigned at least one, here two wheel sets 24 . The bogie 18 is supported on a wheel set bearing housing 28 via a primary suspension system 26 . Such a structure of a chassis of car 10 is generally known, so that this will not be discussed in more detail in the present description.

The function of the primary suspension systems 26 and secondary suspension systems 20 is to dampen impacts occurring on the railroad path during the intended use of the vehicle 10 , so that the body 22 of the vehicle 10 is essentially unaffected by given impacts.

The primary suspension system 26 is associated with an acceleration sensor 30 . The acceleration sensor 30 is in this case connected to a component 32 which is fixedly connected to the wheelset bearing housing 28 and thus to the wheelset bearing. Such an acceleration sensor 30 is provided on at least one wheel set 28 of a bogie 18 . It is preferred before an acceleration sensor 30 is assigned to both wheels of a wheel set 28 . Acceleration sensors 30 have - corresponding to the coordinate system indicated in FIG. 1 - at least one detection direction running in the z direction. Acceleration sensors 30 are preferably used, which have a detection direction in the z direction and the y direction. Acceleration sensors 30 are also conceivable, which can detect acceleration effects in the z-direction, y-direction and x-direction. Here, in the context of the present invention, an acceleration effect in the x direction, that is to say in or opposite to the direction of travel 12 , is of no interest. The z direction is defined vertically to the direction of travel 12 and the y direction is defined transversely to the direction of travel 12 .

The vehicle 10 includes a data acquisition and evaluation unit 31 , which is connected to the at least one acceleration sensor 30 . The data acquisition and evaluation unit 31 is also connected to display devices for a driver of the vehicle 10 and / or to actuators of the vehicle 10 for initiating a braking operation.

Structure and mode of operation of acceleration sensors are generally known, so that in the context of this description also not in more detail should be gone.

By means of the acceleration sensor 30 , an acceleration of the component 32 , and since this is firmly connected to the wheel set bearing (wheel set bearing housing 28 ), an acceleration of the wheels of the wheel set 24 is detected in the z direction. When the vehicle 10 is used as intended, accelerations due to bumps, rail joints, switch junctions, bridge entries or exits, tunnel entries or exits are detected in the z direction. If the wheelset 24 is assumed to be derailed, it suddenly jumps off the rails of the track 14 , so that an additional acceleration takes place at least in the z direction (in the case of lateral deflection also in the y direction). To detect such a derailment, the signal delivered by acceleration sensor 30 is evaluated, as described below. The signal delivered by the acceleration sensor 30 can first be subjected to filtering, in particular low-pass filtering, in the data acquisition and evaluation unit 31 in order to eliminate interference signals caused by driving.

First, a fixed coordinate system is defined. This definition is carried out as entered in FIG. 1 alternatively that at a time t ₀ the wheelset bearing 28 has a distance S ₀ in the vertical direction from the origin of the coordinate system. While the train 10 is traveling in the direction of travel 12 , the wheelset bearing 28 has an arbitrary distance S _{i from} the origin at any time t _i . For this purpose, it is entered in FIG. 1, for example, that the distance S ₁ is at time t ₁ . Due to the rigid coupling of the acceleration sensor 30 with the component 32 (wheelset bearing 28 ) and its known defined distance from the wheelset bearing 28 , the path change ΔS = S ₁ - S ₀ or generally formulated ΔS = S _j - S with the aid of the acceleration sensor 30 _j-1 can be determined. The acceleration sensor 30 is thus in a defined, unchangeable position relative to the point of contact between the wheel and the rail.

These path data resulting in the z direction cannot or cannot be determined directly with the acceleration sensor 30 . It applies here that the information content of acceleration signals is less than that of path signals. While the acceleration can be determined via the mathematical derivation of the path signals, the path can only be determined from the acceleration signals (supplied by the acceleration sensor 30 according to the invention) only by knowing the initial conditions. It was found, however, that reliable derailment detection of the vehicle 10 is possible even without direct displacement measurement by means of the acceleration sensor 30 . Here it is assumed that the relationship between distance, speed and acceleration is derived from the relationship

results. Here, t _{i-1 is} the start time of an integration interval under consideration, t _{i is} the end time of the integration interval, V _{0 is} the speed of the wheelset bearing 28 at the time t _i-1 in the z direction and S _{0 is} the already explained distance of the wheelset bearing 28 to the origin the coordinate system; a is the acceleration of the wheelset bearing 28 in the z direction measured by means of the acceleration sensor 30 ; S _i is the distance of the wheelset bearing 28 from the origin of the fixed coordinate system at the point in time t _i .

Fig. 2 illustrates the relationships between distance, speed and acceleration. The path (distance) S is plotted against the time t. At time t ₀ , the distance S ₀ . At the same time, the speed V _{0 is given} at time t ₀ . The acceleration signal provided by the acceleration sensor 30 and integrated twice in the data acquisition and evaluation unit 31 is denoted by 34 . This runs with fluctuations around a zero line 36 . The dual integrated acceleration signal 34 is accordingly decreasing or increasing, depending on whether the distance of the wheelset bearing 28 to the origin of the fixed coordinate system decreases (S becomes smaller) or increases (S becomes larger). The twice integrated acceleration signal 34 increases exponentially over time t. This is due to inevitable measurement errors and numerical calculation errors as well as geological disturbances that occur in the integration. In order to eliminate these sources of error, a total integration interval Δt = t ₁ - t _{0 is} defined for the evaluation of the twice integrated acceleration signal 34 , as will be explained below.

In FIG. 2, a boundary G is _s supported a the same time, which is defined by a lower limit and an upper limit G _u G _o. These limit values G _u and G _o define a distance S of the wheelset bearing 28 from the origin of the stationary coordinate system, detection of a derailment if the limit values are undershot and / or exceeded. This is possible because in the event of derailment, the wheel of the wheelset 24 and thus the wheelset bearing 28 comes off the rail 14 and thus shifts in the z-direction of the coordinate system. This leads to a change in the distance S and an acceleration of the wheelset bearing 28 in the z-direction. This acceleration can be evaluated by means of the acceleration sensor 30 .

Provision can also be made to detect a bandwidth of the upper and lower reversal point of the twice integrated acceleration signal 34 and to detect on the track when a predeterminable bandwidth is exceeded.

The formula for determining the distance S _{i of} the wheel set bearing 28 from the origin at the time t _i includes, as unknown, the speed V ₀ and the distance S _{0 of} the wheel set bearing 28 at the time t ₀ . However, these initial conditions are not relevant for the consideration of a derailment of the vehicle 10 , since, as already explained, the change in route AS at time t _i compared to time t _i-1 or t ₁ compared to t ₀ is decisive. It is thus clear that a change in the path ΔS can be concluded by double integration of the acceleration signal 34 . The speed V ₀ at time t ₀ , which is also unknown, can be separated in a simple manner on the basis of its known linear component in the twice integrated acceleration signal 34 . Here, for an angle α, which is spanned by the time axis and the regression line of the twice integrated acceleration signal 34 , V ₀ = tan (α). By determining the angle α, the speed V ₀ can thus be deduced, so that the tendency of the signal curve of the twice integrated acceleration signal 34 can be set to 0.

Based on the explained evaluation of the twice integrated acceleration signal 34 , the path change ΔS of the wheelset bearing 28 between two successively defined times t ₀ and t ₁ , which form an integration interval Δt, can be determined. From the relationships already explained, this clearly shows whether the distance S of the wheelset bearing 28 at time t _{1 has} exceeded or fallen below the limit value G _u or G _o and / or has exceeded the predetermined bandwidth, so that it knows of a derailment can be.

Fig. 3 illustrates the definition of the integration intervals At. The integration intervals Δt are to be determined in such a way that on the one hand a complete possible derailment process is recognized and this is however recognized so quickly that a corresponding action, for example stopping the vehicle 10 or the like, can be initiated immediately after the derailment.

Referring to Figs. 3 and 4, the fixing of the integration interval .DELTA.t is to be clarified. Fig. 3a shows here first the course of the twice integrated acceleration signal 34 over time t. The integration interval is determined by the time t ₀ and the time t ₁ . Here, the course of the double integrated acceleration signal 34 initially still has a relatively flat rise. This results from the fact that the time t _{0 is defined} as the start of evaluation of the double integrated acceleration signal 34 . . Is the Inte grationsintervall .DELTA.t of FIG ends 3a, the time t ₁ as a new time t ₀ - as FIG. 3b illustrates - defined, so that a new Integra t tion interval .DELTA.t from time ₀ to time t ₁ starts. Due to the redefinition as time t ₀ , the twice integrated acceleration signal 34 is again in the region of a relatively flat increase. Continues, as is indicated more in Fig. 3c, the respective time t ₁ for the subsequent integration of the acceleration signal 34 as a new time t ₀ is set. As a result, as indicated in FIG. 4a, there is a series of integration intervals Δt between the successive times t ₀ and t ₁ . The length of an integration interval .DELTA.t is selected so that, on the one hand, the disturbing influences already mentioned on the acceleration signal 34 are eliminated. This is done by redefining the time t ₀ . Furthermore, the integration interval Δt must be so long that a complete Entglei solution process can be reliably detected. The length of the integration interval Dt is, for example, between 0.1 and 1.0, in particular between 0.2 and 0.5 sec.

Since the actual time t _{i of} a possible derailment of the vehicle 10 cannot be predicted, the integration intervals Δt are overlapped in accordance with the preferred variant illustrated in FIG. 4b. Here, a time-shifted evaluation of the double integrated acceleration signal 34 begins. The integration intervals Δt overlap between 0.2 and 0.5 Δt, for example 0.5 Δt. This ensures that derailment of vehicle 10 is reliably detected at any time.

The evaluation of the twice integrated acceleration signal 34 of an acceleration sensor 30 was explained on the basis of the preceding figures. In order to increase the security of the derailment detection, it can be provided that the acceleration signals 34 are evaluated by two acceleration sensors 30 . These can be, for example, acceleration sensors 30 arranged on opposite wheel set bearings of a wheel set 24 . If appropriate, the acceleration sensors 30 can also be arranged on wheelsets 24 which follow one another in the direction of travel 12 . By comparing the two acceleration signals, these can be verified, as operational interference can be eliminated. For example, in the direction of travel 12 on successively arranged acceleration sensors 30, for example a turnout can be recognized by shock events which continue in the longitudinal direction of the vehicle. In the event of derailment of a vehicle 10 , both wheels of a wheel set 24 and / or the wheels on successive wheel sets are normally derailed, so that the path change and acceleration acting in the z direction occur on these. If such a change in path or acceleration occurs only on one wheel, it can either be concluded that there is an operational interference which does not constitute derailment. On the other hand, when evaluating the twice integrated acceleration signal 34 of an acceleration sensor 30 , a broken wheel, broken tire or the like can also be detected. Such a wheel break does not necessarily mean an immediate derailment of the vehicle 10 . However, a path change ΔS then occurs unilaterally, which falls outside the specified bandwidth G _s , so that an intolerable interference can be concluded, so that appropriate countermeasures, for example stopping the vehicle 10 , can be initiated.

Claims (17)

1. Method for detecting a derailment track-bound vehicle, with an acceleration one with the lane directly or indirectly in Contact standing component of the track-bound Vehicle vertically and / or across a journey Direction is determined and the acceleration is measured with an acceleration sensor and a signal delivered by him twice over time is integrated and this double integrated Acceleration signal with an upper and / or lower limit (Gu, Gs) is compared and at Falling below and / or exceeding the limit is recognized on a derailment. 2. The method according to claim 1, characterized in that that a bandwidth between an upper and lower Reversal point of the double integrated acceleration supply signal with a predeterminable bandwidth is compared. 3. The method according to claim 2, characterized in that that the acceleration signal to eliminate filtered signal components due to driving, in particular is low pass filtered.   4. The method according to any one of claims 2 or 3, characterized in that from the double integrated acceleration signal, a path change (ΔS) of the component directly or indirectly in contact with the lane to a defined origin of a coordinate system at successive times (t _i ) and (t _i-1 ) or (t ₁ ) and (t ₀ ) is determined. 5. The method according to any one of claims 2 to 4, characterized in that the evaluation of the two fold integrated acceleration signal within a definable integration interval (Δt) follows. 6. The method according to any one of claims 2 to 5, characterized in that for determining the path change (ΔS) within an integration interval (Δt), the acceleration signal supplied by the acceleration sensor is integrated twice over time and an output speed (V ₀ ) decreasing portion of the double integrated acceleration signal is eliminated. 7. The method according to claim 6, characterized in that the integration intervals (Δt) are determined successively, the end time (t ₁ ) of an integration interval (Δt) always forming the start time (t ₀ ) of the next integration interval (Δt). 8. The method according to claim 6 or 7, characterized records that the integration intervals (Δt) over start lapping with a time lag. 9. The method according to claim 8, characterized in that an overlap of the integration intervals is between 0.2 and 0.5 Δt. 10. The method according to any one of claims 6 to 9, there characterized in that a length of Integration intervals (Δt) between 0.1 and 0.5 Seconds, in particular 0.2 seconds. 11. The method according to any one of claims 2 to 10, characterized in that the acceleration signals from at least two acceleration sensors be compared with each other. 12. Device for detecting derailment of a track-bound vehicle, characterized by an acceleration sensor ( 30 ) which is assigned to a component of the track-bound vehicle ( 10 ) that is in direct or indirect contact with the lane ( 16 ), the acceleration sensor ( 30 ) is connected to a data acquisition and evaluation unit ( 31 ) which integrates the signal ( 34 ) delivered by the acceleration sensor ( 30 ) twice over time, and compares the processed signal with an upper and / or lower limit value (Gu, Gs) and an action can be triggered if the limit value is undershot and / or exceeded (Gu, Gs). 13. The apparatus according to claim 12, characterized in that the acceleration sensor ( 30 ) in the loading area of the wheelset storage is assigned a primary suspension system ( 26 ) of the vehicle ( 10 ). 14. Device according to one of claims 12 or 13, characterized in that the acceleration sensor ( 30 ) is arranged on a wheelset bearing housing. 15. Device according to one of claims 12 to 14, characterized in that an acceleration sensor ( 30 ) is assigned to each wheel of a wheel set ( 24 ). 16. The device according to one of claims 12 to 15, characterized in that the acceleration sensors ( 30 ) have at least one vertical to a direction of travel ( 12 ) direction of detection (z-direction). 17. Device according to one of claims 12 to 16, characterized in that the acceleration sensors ( 30 ) have a vertical and a transverse to the direction of travel ( 12 ) direction of detection (z-direction, y-direction).