EP1609691B1 - Method of minimizing the wheel wear of a railway vehicle - Google Patents

Abstract

The railway wheel slipping minimization system aims to keep each axle radially aligned when the vehicle (SCH) is negotiating a curve. Several sensors (SE1) transmit data (DAT) to an evaluation circuit (ASW). The computer calculation system may involve fuzzy logic. The evaluation circuit produces a desired value signal (PAO) which is passed to a control circuit (STR) and a regulating circuit (REG). A further sensor (S2) is connected to the regulating circuit. The regulating circuit is connected to a drive or braking system for the individual wheels (ANT). The control circuit is connected to an actuator (STG) for swiveling each individual axle. Calculations may be made using coefficient of friction, vehicle speed and relative speed of separately rotating wheels.

Description

The invention relates to a method for minimizing the wheel wear of a rail vehicle.

Known methods for minimizing the friction loss and thus the wheel wear in the wheel / rail contact in rail vehicles try during a bow ride to influence the position of the wheels relative to the track so that sliding effects are avoided or minimized at the contact point. In all known methods, therefore, the supposedly optimal position should be known in advance.

The majority of rail vehicles in use today are equipped with wheelsets (more or less rigid coupling of the wheel disks via the wheelset shaft). In such vehicles it is assumed that the optimum position of the wheels in the track corresponds to the radial position. This can be found among others in the EP 0 600 172 A1 . DE 30 04 082 A1 . EP 0 007 225 A1 ,

In the from the EP 0 600 172 A1 Although known method sets in many operating conditions a favorable wear behavior, but this does not correspond to the optimum. Rather, the friction optimum for wheel sets depends on a number of factors, such as vehicle speed, friction coefficient, vehicle geometry, radius of curvature of the tracks, wheel / rail geometry, etc. In general, the friction optimum does not correspond to the radial position but to a position in the track in which both longitudinal and transverse slips occur.

In addition, in the from the EP 0 600172 A1 Known methods realized no radial position of the wheelsets relative to the track, but only the angle between the wheelset and chassis frame adjusted according to the radial position. However, the actual position of the wheels relative to the track remains unknown in this method.

The situation is similar for vehicles with single wheels.

Although there corresponds to the radial position of the wheels as long as the friction optimum, as long as due to the wheel / rail contact geometry no multipoint contact occurs. However, if a multipoint contact occurs when the wheels are in a radial position, high longitudinal slippage occurs. The actual optimum wheel position, in turn, then depends on a number of factors, such as vehicle speed, coefficient of friction, vehicle geometry, rail bend radius, wheel / rail geometry, etc.

The known methods, which relate to the friction minimization in rail vehicles with single wheels, generally assume that the radial position of the wheels in the track always corresponds to the friction power optimum.

So revealed the DE 198 26 452 a method for the drive coordination of a single-wheel drive, track-guided vehicle when bow. The vehicle has at least one axis rotatably mounted relative to the chassis frame, wherein in a variant of the known embodiment, the angular position of the axle relative to the chassis frame is controlled so that the wheels are tangential to the rail (= "radial position"). The regulation is carried out by adjusting the angular position of the axis via the torque setpoints of the individual wheel drives. Although this method provides more favorable wear behavior in many operating states, the achievable wear behavior is generally not optimal here as well. In addition, also in this method, only the angle of the wheel axles is generated relative to the chassis frame corresponding to the radial position. The actual position in the track remains unknown even in this procedure.

From the DE 199 10 256 A1 is an active radial control for a rail vehicle in the track arc known. A computing unit outputs control signals to active actuators on the output side as a function of the track data received on the input side.

The DE 197 02 409 A1 describes a method for detecting the position of track-guided vehicle wheels in the track channel by measuring the local torque-dependent preforming of the bearing environment as an element of a single-wheel or wheelset control.

It is therefore an object of the invention to provide a way which makes it possible to further reduce wheel wear on a rail vehicle and thus increase the life of the wheels.

This object is achieved by a method according to claim 1.

Variables that are relevant for the friction performance represent, for example, parameters such as routing data (radius of curvature, track cant, etc.), wheel / rail contact ratio (sliding friction coefficient for the wheel / rail contact, contact geometry, etc.), vehicle speed, payload mass and drive or braking torques.

Parameters for an optimal position describing a minimum friction loss of the wheels are, for example, a transverse deflection of the axles relative to the track, the starting angle relative to the track (= yaw angle relative to the track) or else a rotational speed difference between the wheels of an axle (in the case of individual wheeled vehicles).

It is an advantage of the invention that the actual optimum position of the wheels in dependence on the relevant data, such as routing data, data on wheel / rail contact ratios, vehicle speed, as well as data on driving and braking torques or the payload of the rail vehicle can be determined and adjusted , whereby the wheel wear can be substantially minimized.

In order to determine the position of the at least one wheel corresponding to the minimum friction power during a curved journey, a mathematical model for a quasi-stationary arc run of the rail vehicle, wherein the model contains at least one variable that is relevant for the friction power, minimizes an equation for the friction power and evaluated with the instantaneous values of the variable quantities.

Furthermore, the wheel position in the track can be calculated in such a way that the friction power is minimized with regard to a profile service life ( Fig. 2 ).

In an easily implemented variant of the invention, depending on the radius of curvature of the tracks, an angle between a wheel axle of the at least one wheel and a chassis frame can be adjusted by means of control.

Furthermore, in single-wheel vehicles as a control variable for the regulation of the wheel position, a differential torque between the wheels of an axle superimposed on the drive and braking torques can be provided. For this purpose, a speed difference between the wheels of an axle can be measured and regulated by means of the differential torque to a calculated target speed difference out, wherein the target speed difference is advantageously calculated assuming cylindrical wheels. In principle, however, the target speed difference can also be calculated assuming any wheel / rail profile pairings. In particular, the target speed difference for a bow travel can be calculated so that the different ways are balanced slip-rolling off, bow outer and inner bow wheels.

According to a further variant of the invention, the transverse deflection of the wheel axle of the at least one wheel can be measured relative to the tracks and be controlled to a calculated Sollquerauslenkung out in single-wheel vehicles. Thus, the transverse deflection can be regulated, for example, by means of a differential torque between the wheels of an axle which is superimposed on the drive and braking torques to the nominal transverse deflection determined by calculation. However, the transverse deflection can also be controlled by means of a steering torque about the vertical axis of the axis to the calculated Sollquerauslenkung out. A particularly easy-to-implement variant of the invention provides that the desired transverse deflection is regulated to the value zero.

Fig. 1

ein Blockdiagramm des erfindungsgemäßen Verfahrens und

Fig. 2

Profile mit unterschiedlichem Materialabtrag.

The invention together with further advantages will be explained below with reference to some non-limiting embodiments, which are illustrated in the drawing. In this show schematically:

Fig. 1

a block diagram of the method according to the invention and

Fig. 2

Profiles with different material removal.

According to Fig. 1 can in the inventive method data DAT the variable, relevant for the friction size or size, such as routing data, data on wheel / rail contact ratios, vehicle speed data, as well as data on driving braking torques of a rail vehicle SCH are recorded while driving, for example by means of suitable sensors SE1.

Another possibility is that the variable size data DAT used for calculation is not picked up during a drive, but is, for example, recorded. B. be queried depending on the route from a database. For example, routing data of predefinable routes can be stored in a database. In principle, the detection of routing data by means of a navigation or location system is possible, for example by means of a satellite-based navigation or location system, such as the GPS system.

The data DAT are forwarded to an evaluation unit ASW, which determines the optimum position of the wheels PAO with regard to the friction power from the data DAT on the basis of an algorithm which contains a mathematical model which describes the interaction of the rail vehicle SCH with the plug.

The algorithm is based on a mathematical modeling of the rail vehicle, which generally corresponds to the equations of motion of the rail vehicle SCH. The mathematical modeling is in the simplest case linear, but can in principle also be nonlinear. For the mathematical modeling of rail vehicles SCH with wheelsets or idler wheels see for example: Kalman Filter for the State Estimation of a 2 Axle Railway Vehicle; TX Mei (UK), RM Goodall (UK), H. Li (UK), European Control Conference 99, Karlsruhe, Germany, August Sept., and "State Estimaton for Active Steering of Railway Vehicles"; H. Li and RM Goodall 1999 IFAC or also the lecture script "Lateraldynamik von Schienenfahrzeugen" by Prof. Dr.-Ing K. Knothe and the DE 4309183 A1 ,

Based on the mathematical model of the rail vehicle SCH, the equation for the frictional power is determined for dynamic or quasi-stationary states in the arc.

wobei P_1,2 die Reibleistung in den Radschienekontakten eines ersten und eines zweiten Radsatzes, f₁₁ und f₂₂ den Kraftschlussbeiwert längs bzw. quer zum Gleisverlauf, λ die Konizität der Räder, r den Radradius, y_1,2 die Querverschiebung der Radsätze relativ zum Gleis, R den Bogenradius eines durchfahrenen Gleisbogens, b die halbe Spurweite, v die Fahrzeuggeschwindigkeit und ψ_1,2 die Verdrehung der Radsätze relativ zu einem Gleis (=Gierwinkel, bzw. Anlaufwinkel) bedeuten. Natürlich kann anstelle der quasistationären Reibleistung, wie bereits oben erwähnt, auch die dynamische Reibleistung betrachtet werden.For linearized rigid-wheeled vehicles, for example, the quasi-stationary friction power may take the following form $$P_{1.2}=2f_{11}{\left(\frac{\lambda }{r}{y}_{1.2}-\frac{b}{R}\right)}^{2}v+2f_{22}{\left({\psi }_{1.2}\right)}^{2}v.$$

where P _{1,2 is} the friction power in the wheel rail contacts of a first and a second wheelset, f ₁₁ and f ₂₂ the adhesion coefficient along or transverse to the track, λ the conicity of the wheels, r the wheel radius, y _1,2, the transverse displacement of the wheelsets relative to the track, R the radius of curvature of a track curve traversed, b half the track width, v the vehicle speed and ψ _1,2 the rotation of the wheel sets relative to a track (= yaw angle, or start-up angle). Of course, instead of the quasi-stationary friction, as already mentioned above, the dynamic friction power can also be considered.

The friction coefficient equation includes, directly and indirectly, depending on the model depth, magnitudes derived from the mathematical model of the vehicle's interaction with the track, such as: B. Vehicle geometry data, stiffness, mass data and the variables mentioned - indirectly therefore da y _1.2 and ψ _1.2 also depend on the mentioned sizes.

The friction energy equation is minimized to determine the friction power optimum. An optimal wheel position is determined by the calculated position of a considered wheel at the determined friction power minimum depending on the instantaneous values of the variable quantities DAT.

und $\frac{\partial P}{\partial T2}=0$

erfolgen, worin T1 und T2 auf die Radsätze wirkende Lenkmomente bedeuten, die die Stellung y_1,2 und ψ_1,2 der Radsätze und somit die Reibleistung P_1,2 beeinflussen und somit Variablen der Funktionen y_1,2 sowie ψ _1,2 darstellen.The minimization of the above equation for the frictional power may, for example, according to the boundary conditions $\frac{\partial P}{\partial T1}=0$

and $\frac{\partial P}{\partial T2}=0$

take place, wherein T1 and T2 acting on the wheelsets steering moments that affect the position y _1.2 and ψ _{1.2 of} the wheelsets and thus the friction P _1.2 and thus variables of the functions y _1.2 and ψ _1.2 represent.

This embodiment of the invention leads, for example, to an optimal steering torque T _opt with respect to the friction power, according to which a friction-optimized control can take place.

Furthermore, the operating point (wheel position in the track) can be calculated in such a way that the friction loss is minimized with regard to the service life of the profile, resulting in uniform wear over the profile ( Fig. 2 ). In Fig. 2 PR1 corresponds to an unworn profile, PR2 to a removal of material at a friction power minimum in terms of profile life, and PR3 to a removal of material at the friction power minimum.

Depending on the embodiment of the invention, the output variable of the algorithm can be, for example, a desired transverse deflection of the axes relative to the track, a nominal start angle relative to the track (= yaw angle relative to the track) or even a nominal rotational speed difference between the wheels of an axle. The setting of the wheel position can be done by means of control or regulation or by a combination of both. To control, for example, corresponding to the wheels of the rail vehicle SCH associated with a control unit STR actuators STG may be arranged. For regulating a control unit REG can be provided. The control can be designed, for example, as state control with or without observer, as control with output feedback or as fuzzy logic control.

In a single wheel vehicle arises from the model of the interaction vehicle rail, that for a large range of possible operating conditions that position of the wheels relative to the track is close to the frictional optimum, where at small starting angles, the transverse deflections of the axles compared to a symmetrical position in the track only very small values. This makes it possible to dispense with the detection of the wheel / rail contact ratios and the payload, the rail geometry and the sliding friction coefficient. This results in the advantageous variants of the invention explained below.

A particularly favorable embodiment of the invention consists in a combination of control and regulation. The control here is that in a vehicle with Einzelradachsen the angle between the Einzelradachsen and the chassis or vehicle frame by the actuators STG, which act from the chassis or vehicle frame on the Einzelradachsen be set. In order to realize a control combined with the control, at least one independent wheel per axle may have its own drive ANT or a brake, whereby it is possible to generate a differential torque between opposing individual wheels via the control unit REG so that an optimal position the Einzelradachsen can be realized in the track. This differential torque is superimposed on the drive torque in driven systems.

To realize the control, the actual speed difference of the wheels of an axle can be measured and the difference to the desired speed difference can be determined. The determined difference represents the input variable of the controller, which regulates the difference over the above-mentioned differential torque to zero. The target speed difference is in the straight track zero, whereby a one-sided start of the wheels is prevented. In the curved track, the target speed difference is calculated, for example, so that the different paths of the slip-rolling outside and bow inner wheels are compensated assuming cylindrical wheels as a function of the vehicle speed, the wheel radius and the radius of curvature. This has the effect that in the curve due to the real conical wheel profiles, the control always leads the Einzelradachsen towards track center, because only there the individual wheels of an axle the same rolling radius - as in the assumed for determining the target speed difference cylindrical wheels - have. In this way, multipoint contact is avoided and the system moves together with the above control in the vicinity of the friction power optimum. The controller is designed so that sufficient stability is ensured for all operating conditions.

In a further variant of the invention, sensors can additionally be provided which detect the transverse deflection of the individual wheel axles relative to the track. This transverse displacement represents the input variable of the controller, which regulates the transverse deflection to the setpoint via the described differential torque. This setpoint is in the straight track zero which prevents unilateral starting. In the curved track, the desired value of the transverse deflection, for example, also be zero, which again a multi-point contact is avoided and together with the control described above, an operation in the vicinity of the friction power optimum can be realized. The controller is again designed so that a sufficient stability for all operating conditions is guaranteed.

Claims (13)

1. Method for minimising the wheel wear of a rail vehicle (SCH), wherein on the basis of data (DAT) for at least one quantity relevant to friction which is variable during a journey of the rail vehicle (SCH), at least one desired value (PAO) is mathematically calculated, corresponding to a minimum friction between at least one wheel and one track, of a parameter characterising the position of the at least one wheel relative to the track, with the wheel position being adjusted according to the desired value by means of an open-loop control, closed-loop control or a combination of both,
   characterised in that
   the position of the at least one wheel corresponding to the minimum friction between the at least one wheel and the track is calculated using a mathematical model of the rail vehicle (SCH) describing the interaction between the rail vehicle (SCH) and the track, and also of the data (DAT) of the at least one variable quantity relevant to the friction.
2. Method according to claim 1, characterised in that to determine the position, corresponding to the minimum friction, of the at least one wheel when running on a bend, an equation is obtained from a mathematical model, containing at least one variable for the friction-relevant quantity, for a quasi-steady-state bend run of the rail vehicle (SCH), is minimised and is evaluated using the momentary values of the variable quantities (DAT).
3. Method according to one of claims 1 to 2, characterised in that the wheel position on the track is calculated so that the friction is minimised with regard to a service life of the profile (Figure 2).
4. Method according to one of claims 1 to 3, characterised in that by means of an open-loop control system an angle is set between one wheel axle of the at least one wheel and an undercarriage frame, depending on the bend radius of the track.
5. Method according to one of claims 1 to 4, characterised in that with individual-wheel vehicles a differential torque, superimposed on the drive and brake torques, is provided between the wheels of an axle as a correcting variable for the closed-loop control of the wheel position.
6. Method according to claim 5, characterised in that a speed differential between the wheels of an axle is measured and corrected to a mathematically determined desired speed differential by means of the differential torque.
7. Method according to claim 6, characterised in that the desired speed differential is calculated assuming any wheel/rail profile pairs.
8. Method according to claim 6, characterised in that the desired speed differential is calculated assuming cylindrical wheels.
9. Method according to claim 7 or 8, characterised in that the desired speed differential for running on a bend is calculated so that the different paths of slip-free rolling outer and inner wheels on the bend can be balanced.
10. Method according to one of claims 1 to 4, characterised in that for individual-wheel vehicles the transverse displacement of the rear axle of the at least one wheel is measured relative to the tracks and corrected to a mathematically determined desired transverse displacement.
11. Method according to claim 10, characterised in that the transverse displacement is corrected to the mathematically calculated desired transverse displacement by means of a differential torque between the wheels of an axle, superimposed on the drive and brake torques.
12. Method according to claim 10, characterised in that the transverse displacement is corrected to the mathematically determined desired transverse displacement by means of a steering torque about the vertical axis of the axle.
13. Method according to claims 10 to 12, characterised in that the desired transverse displacement is corrected to the value zero.

Images