Method and device for monitoring an electric network in a rail vehicle and
rail vehicle
ln railvehicles, inter a/ra, so-called permanent magnet machines, which can also be
referred to as permanent magnet motors, are used to drive the rail vehicle. They are
generally supplied with electric energy via a three-phase electric network. The electric
network also comprises a power converter in this case, which, in motor operation of the
permanent magnet machine, converts a DC voltage, for example, a link circuit voltage,
into a desired AC voltage for supplying the permanent magnet machine. The power
converter can also, in generator operation of the permanent magnet machine, convert the
AC voltage generated by the permanent magnet machine into a DC voltage, however.
Undesired short-circuits can occur in the electric network. They can occur both inside the
permanent magnet machine, for example, inside a housing of the machine, or along a
phase line for connecting the power cenverter and the permanent magnet machine. Shortcircuits
can also occur in the power converter. In the event of short-circuits, so-called
electric arcs can also occur, which can result in undesired destruction of components of
the railvehicle.
Monitoring a levelof a phase current is known, wherein a shortcircuit is detected if the
level of the phase current exceeds a predetermined threshold value.
lf such a short circuit is detected, in generalthe power converter is electrically
disconnected from the permanent magnet machine, for example, by appropriately
arranged circuit breakers. At the same time, the rail vehicle is decelerated to a standstill,
to prevent a potential feed of the short-circuit.
The technical problem arises of providing a method and a device for monitoring an
The technical problem arises of providing a method and a device for monitoring an electric
network in a rail vehicle and also a rail vehicle, which enable monitoring of the network
even in the disconnected state of the electric machine.
The solution to the technical problem results by way of the subject matter having the
features of claims 1 , 10, and 12. Further advantageous embodiments of the invention
result from the dependent claims.
A method is proposed for monitoring an electric network in a rail vehicle. The eleciric
network can be in particular a traction network of the rail vehicle or can refer to a part of
the traction network of the rail vehicle. The electric network is used in particular for energy
transmission between a power conveder of the railvehicle and a permanent magnet
machine of the rail vehicle.
The electric network comprises at least one power converter. The power converter can be
operated in this case both as an inverter and also as a rectifier. For example, the power
converter can be designed as a three-phase power converter. The power converter can
comprise in pafiicular electric switching elements, for example, MOSFET or IGBT.
The power converter can be electrically connected on the input side to a link circuit, in
particular a link circuit capacitor, of the rail vehicle. A link circuit voltage which drops via
the link circuit capacitor, and which therefore is applied on the input side to the power
converter, is a DC voltage. On the output side, the power converter can be connected to
the permanent magnet machine, for example, via at least one phase line.
Furthermore, the electric network comprises the at least one permanent magnet machine.
The permanent magnet machine refers in this case to a synchronous machine having a
permanently magnetized rotor. The permanent magnet machine can be operated in motor
operation. ln this case, electric energy which is transmitted from the power converter to
the permanent magnet machine is converted into mechanical energy. The electric energy
is transmitted in this case in the form of an alternating current and an AC voltage, which
feed the permanent magnet machine. ln a generator operating mode, mechanical energy
is converted by the permanent magnet machine into electric energy, wherein the electric
energy can be transmitted to the power converter. ln this case, the permanent magnet
machine generates an alternating current and an AC voltage.
Furthermore, the electric network comprises at least one first phase line for the electrical
connection of the at least one power converter and the at least one permanent magnet
machine. The first phase line refers in this case to an electric line through which a first
phase current can flow. The electric network preferably comprises more than one, in
particular three phase lines. At least one electric switch element, for example, a power
switch element, in particular a MOSFET, an IGBT, or a circuit breaker, can be arranged
along the phase line. By means of the electric switch element of the first phase line, an
electrical connection of the power converter and the permanent magnet machine via the
first phase line can be interrupted or established.
The power converter is preferably a three-phase power converter, which is connected via
three phase lines to a three-phase permanent magnet machine.
Furthermore, the first phase line is interrupted. For this purpose, as will also be explained
in greater detail hereafter, a corresponding interruption unit can be actuated. The
interruption unit is preferably designed as an electric switch element. The electric switch
element can be arranged along the first phase line or in the first phase line. A connection
of the power converter and the permanent magnet machine via the first phase line can be
established or interrupted by means of the electric switch element. For example, the first
phase line can be interrupted in that the electric switch element is opened.
Furthermore, a potential difference is determined between a machine-side part of the
phase line and a reference potential. The reference potential can be, for example, a
referencing potential, for example, a ground potential of the electric network. Alternatively,
the reference potential can be the potential of a further phase line of the electric network,
in particular of a machine-side part. of this further phase line.
The machine-side part of the phase line refers in this case to a part or section of the
phase line which connects a determination point of the potential difference to the
permanent magnet machine. The determination point in turn refers to a point at or in
which the potential difference is determined. The determination point can be equivalent to
the interruption point, for example. The term "point" can also refer to a partial section of
the phase line in this case, of course.
The determination of the potential difference comprises in this case the direct detection of
the potential difference, for example, by means of a suitable detection unit, in particular by
means of a voltage sensor. However, the term "determination" can also comprise the
calculation of the potential difference as a function of at least one directly detected
intermediate variable.
Furthermore, a potential-difference-dependent variable is determined. The potentialdifference-
dependent variable forms in this case an indicator of a level of the potential
difference. Exemplary potential-difference-dependent variables will also be explained in
greater detail hereafter.
Furthermore, a speed of the permanent magnet machine is determined. The speed can
be detected directly in this case, for example, by means of a suitable detection unit, in
particular a speed sensor. However, it is also possible to determine the speed by
computation as a function of at least one directly detected variable, for example, a travel
velocity of the rail vehicle. Of course, still further, in particular previously known,
parameters of the rail vehicle can also be taken into consideration for this purpose, for
example, transmission ratios of a mechanical connection of the permanent magnet
machine and rotatable wheels of the rail vehicle.
Furthermore, a speed-dependent reference variable, which can also be referred to as a
speed-dependent variable, is determined as a function of the speed. ln this case, a
relationship, for example, a functionalrelationship, between the speed and the speeddependent
variable can be previously known. The speed-dependent variable refers in this
case to a variable corresponding to the potential-difference-dependent variable. ln
particular, the speed-dependent reference variable can form an indicator of a level of an
idle voltage of the permanent magnet machine at a specific speed.
For example, it can be assumed that a linear relationship exists between the speed and
the level of the idle voltage, wherein the idle voltage increases with rising speed.
Furthermore, a deviation of the potential-difference-dependent variable from the speeddependent
reference variable is determined. ln particular, a difference can be determined
between the speed-dependent reference variable and the potential-difference-dependent
variable.
Furthermore, a network fault is detected if the deviation is greater than a predetermined
threshold value. ln particular, a network fault can thus be detected if the previously
explained difference or an absolute value of the difference is greater than the
predetermined threshold value. The predetermined threshold value can be zero in this
case or can be a predetermined small amount greater than zero.
ln this way, a network fault in a machine-side partial network, in parlicular in the
permanent magnet machine, can advantageously be detected when the permanent
magnet machine is electrically disconnected from the power converter. The machine-side
partial network refers in this case to at least the part of the electric network which is
arranged between a determination point of the electric network and the permanent
magnet machine, wherein the permanent magnet machine itself is also part of the
machine-side partial network. The machine-side partial network can therefore comprise at
least the section of the first phase line which connects the above-explained determination
point to the permanent magnet machine and also at least a part of the electric network of
the permanent magnet machine.
Furthermore, a type of fault can be determined as a function of the level of the deviation,
for example, the level of the above-explained difference. lf a potential difference of zero or
nearly zero is provided, for example, and therefore a high, in particular maximum,
deviation of the potential-difference-dependent variable from the speed-dependent
reference variable is provided, a short-circuit can thus be detected. However, if the
determined potential difference is not equal to zero, but is less than the expected idle
voltage, a deviation, but not a maximum deviation, of the potential-difference-dependent
variable from the speed-dependent reference variable therefore exists. ln this case, for
example, an undesired low-impedance connection can be detected.
For example, a type of fault can be assigned to a predetermined deviation interval. lf the
deviation determined according to the invention is in one of the fault-dependent deviation
intervals, the fault which is assigned to the corresponding fault-dependent deviation
interval can thus be detected as the type of fault.
The method according to the invention can in particular only be carried out if the first
phase line is interrupted. This can bethe case, forexample, íf the phase line is already
interrupted because of a further network fault, in particular a network fault in the power
converter. An interruption can also be provided if a desired non-operation of the motor, for
example, to save energy, is provided, wherein in this case the first line can also be
interrupted. ln the case of non-operation of the permanent magnet machine, for example,
a further permanent magnet machine of the rail vehicle may take over the drive.
ln a further embodiment, the electric network comprises three phase lines. Furthermore,
at least two, in particular two of the three, or all phase lines are interrupted, and a potential
difference between a machine-side part of the phase line and a reference potential is
determined for each phase line. Furthermore, a potential-difference-dependent variable is
determined in each case.
For this purpose, a potential difference between the phase line and a common reference
potential, for example, a ground potential of the electric network, can be determined for
each phase line. Alternatively and preferably, however, the potential of one of the
remaining phase lines forms the reference potential for each phase line. This enables, as
will also be explained in greater detail hereafter, only two voltage sensors to be used, for
example, to determine a potential difference for each of three phase lines. ln this case, the
voltage sensors can each detect a potential difference between phase lines of a line pair,
wherein line pairs differ in each case in at least one line of the two lines of the line pair.
Furthermore, as explained above, a speed of the permanent magnet machine and, as a
function of the speed, a speed-dependent reference variable is determined for each phase
line.
ln a first alternative, a deviation of the potential-difference-dependent variable from the
speed-dependent reference variable is determined for each phase line and a network fault
is detected if at least one of the deviations is greater than a predetermined threshold
value.
Alternatively or cumulatively, a minimum potential difference of all potential differences
and a variable dependent on this minimum potential difference are determined. The
variable dependent on the minimum potential difference is determined in this case like the
above-explained potential-difference-dependent variable. Furthermore, a deviation of this
variable, which is dependent on this minimum potential difference, from the speeddependent
reference variable is determined and a network fault is detected if this
deviation is greater than a predetermined threshold value.
This advantageously has the result that a network fault can also be determined in a threephase
electrical network.
ln a further embodiment, a two-phase network fault or a three-phase network fault is
determined as a function of the potential differences and/or the at least one deviation.
lf at least one of the deviations is greater than the predetermined threshold value and if at
least one of the potential differences deviates from the remaining potential differences by
more than a predetermined amount, a two-phase network fault can thus be detected.
Such a two-phase network fault can be a short-circuit between two phases, for example.
Such a two-phase network fault can also be detected if at least one deviation is greater
than the predetermined threshold value, but a further deviation is less than or equal to the
predetermined threshold value.
A three-phase network fault, for example, a short-circuit between all phases, can be
detected if at least one deviation is greater than the predetermined threshold value and all
potential differences are equal. A three-phase network fault can also be detected if all
deviations are greater than the predetermined threshold value.
A type of the network fault can advantageously be determined in this way.
ln a further embodiment, a speed of the permanent magnet machine is reduced upon
detection of a network fault. A rotor of the permanent magnet machine can be braked for
this purpose, for example. However, the rail vehicle can preferably be braked by means of
at least one braking unit for this purpose. Of course, the speed of the permanent magnet
machine can be reduced to zero. The reduction of the speed advantageously has the
result that a short-circuit is not still fed by the permanent magnet machine. The risk of
undesired damage can be reduced in this way.
Alternatively or cumulatively, a further interruption of the machine-side part of the phase
line can be carried out. ln particular, the electrical connection between the phase line and
a connection point of the permanent magnet machine can be interrupted, in particular in
the connection point of the permanent magnet machine. lt is also conceivable that the
phase line comprises a further electric switch element, which is arranged in the machineside
part of the phase line. The further interruption can be performed in this case by
opening the further electric switch element.
A fault location, which is located in the machine-side part of the phase line, can be
electrically isolated by a further interruption of the machine-side part of the phase line.
Preferably, an interruption as close as possible to the machine of the machine-side part of
the phase line is carried out. ln particular, a distance of an interruption point from the
connection point of the phase line to the permanent magnet machine can be less than a
predetermined distance.
ln a further embodiment, an accuracy of the determination of the potential difference is
checked. lf, as explained in greater detail hereafier, a determination unit is used for
determining the potential difference, a functionality of this determination unit can thus be
checked in this way. The accuracy of the determination of the potential difference is
checked in that in the case of an uninterrupted phase line, at least one normal operating
potential difference is determined and a deviation of the normal operating potential
difference from a normal operating reference value is determined. A sufficient accuracy is
detected if the deviation is less than or equal to a predetermined threshold value. A
deviation of the normal operating potential difference from a normal operating reference
value can also be determined in particular in that, as a function of the normal operating
potential difference, a potential-difference-dependent variable is determined and, as a
function of the normal operating reference valr:e, a corresponding reference variable is
determined, wherein a deviation of the potential-difference-dependent variable from the
corresponding reference variable is determined.
The normal operating reference value can be equal to the link circuit voltage, for example.
ln particular, the normaloperating potential difference corresponds to a potential
difference to be expected under normal operating conditions, which can be determined
computationally, for example. The accuracy of the determination of the potential difference
can be determined, for example, as a function of present operating parameters of the
electrical network. ln this case, the normal operating potential difference can be
determined as a function of the present operating parameters, for example, as a function
of switch states of the switch elements of the power converter.
It is presumed in this case that the accuracy of the determination of the potential
difference is provided in the case of an interrupted phase line if a correct determination is
performed in the case of an uninterrupted phase line.
Alternatively or cumulatively, the accuracy of the determination of the potential difference
can be checked in that a standstill of the permanent magnet machine is detected, wherein
at least one standstill potential difference is detected in the case of an uninterrupted
phase line, wherein a deviation of the standstill potential difference from a standstill
reference value is determined, wherein a sufficient accuracy is detected if the deviation is
less than or equal to a predetermined threshold value. A deviation of the standstill
potential difference from a standstill reference value can in particular also be determined
in that a potential-difference-dependent variable is determined as a function of the
standstill potential difference and a corresponding reference variable is determined as a
function of the standstill reference value, wherein a deviation of the potential-differencedependent
variable from the corresponding reference variable is determined.
ln general, it is possible that an output signal which represents or codes the potential
difference or the potential-difference-dependent variable is inversely proportional to a level
of the potential difference or the potential-difference-dependent variable. This means that
a high signal level or a high output signal is generated if a low potential difference is
determined. At a standstill of the permanent magnet machine, the potential difference to
be expected has a low value according to expectation, theoretically a value of zero.
Therefore, a high output signal can be expected. However, a lower output signal than the
expected output signal can be generated at a standstill as a result of faults in a
determination unit. Therefore, faults of the determination unit, for example, electronics
faults, can advantageously be detected.
ln a further embodiment, the speed is associated with one speed interval of multiple
speed intervals, wherein an interval-dependent reference variable is assigned to one
speed interval, wherein the speed-dependent reference variable is determined as the
interval-dependent reference variable.
A method which is simple to carry out by computation advantageously results in this way.
ln a further embodiment, the potential difference is assigned to one potential difference
interval of multiple potential difference intervals, wherein an interval-dependent potential
difference value is assigned to one potential difference interval. The interval-dependent
potential difference value can be, for example, a minimum or a maximum value of the
potential difference interval. Furthermore, the potential-difference-dependent variable is
determined as a function of the interval-dependent potential difference value. Carrying out
the method by computation is also advantageously simplified in this way. ln particular,
however, one of the types of faults explained above can be assigned to one or more
potential difference intervals. A simplified classification of a type of fault therefore
advantageously results.
t0
ln a further embodiment, a potential difference between a machine-side part of the phase
line and a reference potential and a potential-difference-dependent variable is determined
for each phase line. Furthermore, a combined output signal is formed, wherein the
combined output signal comprises a first bit sequence, which codes the potentialdifference-
dependent variable formed from the minimum potential difference. This bit
sequence can comprise a predetermined number of bits, for example, 3 bits. Furthermore,
the combined output signal comprises a further bit sequence, which codes an equality of
all potential differences. The further bit sequence can in particular consist of precisely one
bit, wherein one of the two states of the bit, forexample, the state "l" orthe state "0 "
codes an equality of all potential differences.
A simple coding of the output signal advantageously results in this way, which simplifies a
detection of a network fault, on the one hand, and the detection of a three-phase or twophase
network fault, on the other hand.
ln a further embodiment, the potential-difference-dependent variable is a maximum
amplitude or an RMS value of the potential difference during a predetermined
determination duration. The determination duration can be, for example, in an interval of
10 msto 100 ms. The determination duration is preferably 100 ms.
The potential-difference-dependent variable can be provided in this case as a digital
signal. The potential-difference-dependent variable can also be generated as inversely
proportional to the level of the potential difference.
Furthermore, the speed-dependent variable can additionally be determined as a function
of a temperature of the permanent magnet machine. Alternatively or cumulatively, the
speed-dependent reference variable can be determined as a function of an age of the
permanent magnet machine. lt can be presumed in this case that the idle voltage
generated by the permanent magnet machine as a function of the speed is additionally a
function of the operating temperature of the permanent magnet machine and/or a function
of an age of the permanent magnet machine. The proposed method can therefore be
improved further in its accuracy and reliability.
Furthermore, a device for monitoring an electric network in a rail vehicle is proposed. The
electric network can be designed as explained above in this case.
ll
The device comprises at least one evaluation unit, at least one determination unit, and at
least one interruption unit. The first phase line is interruptible by means of the interruption
unit. A potential difference between a machine-side part of the phase line and a reference
potential is determinable by means of the determination unit.
Furthermore, a potential-difference-dependent variable is determinable. The potentialdifference-
dependent variable can be determined in this case, for example, by the
determination unit. The potential-difference-dependent variable can thus be provided, for
example, in the form of an output signal, in particular a digital output signal, of the
determination unit. Alternatively, the potential-difference-dependent variable can also be
determined by the evaluation unit, however.
The evaluation unit and the at least one determination unit have a data and/or signal
connection in this case. The evaluation unit and the determination unit can also be formed
as a joint unit.
Furthermore, a speed of the permanent magnet machine and, as a function of the speed,
a speed-dependent reference variable are determinable. The device can comprise a
speed determination unit in this case.
The determination of the speed-dependent reference variable can also be carried out in
this case by means of the determination unit or by means of the evaluation unit.
Furthermore, a deviation of the potential-difference-dependent variable from the speeddependent
reference variable is determinable by means of the evaluation unit.
Furthermore, a network fault is detectable if the deviation is greater than a predetermined
threshold value.
The device advantageously enables one of the above-described methods to be carried
out. The device is therefore designed such that the method according to one of the aboveexplained
embodiments can be carried out by means of the device.
ln one preferred embodiment, the potential difference is determined close to the motor.
This can mean that a determination point for the determination of the potential difference
is not situated spaced apart by more than a predetermined distance from a connection
t2
point of the phase line to the permanent magnet machine, wherein the distance is
measured along the phase line.
ln this way, a very low probability of an incorrect detection of a network fault as a result of
a fault in the wiring between voltage sensor and permanent magnet sensor
advantageously results.
Furthermore, a rail vehicle is proposed, wherein the rail vehicle comprises a device
according to one of the above-explained embodiments. A rail vehicle advantageously
results in this way, the operational reliability of which is increased.
The invention will be explained in greater detail on the basis of an exemplary embodiment.
ln the figures:
Figure 1 shows a schematic circuit diagram of an electric network of a rail vehicle,
Figure 2 shows a schematic block diagram of a device according to the invention,
and
Figure 3 shows a schematic illustration of speed intervals and voltage intervals,
Hereafter, identical reference signs identify elements having identical or similar technical
features.
Figure 'l shows an electric network 1 of a rail vehicle (not shown). The electric network 1
comprises a power converter C and a permanent magnet machine M, wherein inductors L
of the permanent magnet machine M are shown by way of example. Furthermore, the
electric network 1 comprises a link circuit capacitor C_k. The power converter C is
connected on the input side to a link circuit capacitor C_k.
The power converter C is a three-phase power converter in this case. lt comprises six
electric switch elements S_C, to each of which a freewheel diode D is electrically
connected in parallel. For the sake of comprehensibility, only one electric switch element
S_C and one diode D of the power convefter C are provided with a reference sign.
13
Furthermore, the electric network 1 comprises three phase lines P1, P2, P3. Furthermore,
phase currents l_P1,1_P2, and l_P3, are shown, which illustrate a current flow through
the corresponding phase line P1 , P2, P3.
Furthermore, the electric network 1 comprises a first current sensor CS1 and a further
current sensor CS3, wherein the first current sensor CS1 detects the first phase current
l_P1 and the further current sensor CS3 detects the third phase current l_P3.
By means of the phase lines P1, P2, P3, the power converter C is connected on the
output side to the permanent magnet machine M. The permanent magnet machine M is
therefore a three-phase electric machine.
A first electric switch element S1_P1 and a second electric switch element S2_P1 are
arranged in the first phase line P1. By opening and closing these two switch elements
51_P1, S2_P1, the electrical connection via the first phase line P'l between the power
converter C and the permanent magnet machine M can be established or interrupted. ln a
closed state of the switch elements S'1_P1, S2_P1, the electrical connection is
established in this case. lf at least one of the two switch elements S1_P1, S2_P1 is
opened, the electrical connection is thus interrupted. ln this case, the first switch element
S1_P1 can designate a so-called motor switch. The second switch element S2_P1 can
designate a further motor switch.
Accordingly, a first electric switch element S1_P2, S1_P3 and a second electric switch
element S2_P2, S2_P3 are respectively also arranged in further phase lines P2, P3.
Furthermore, a first voltage sensor VS1 and a second voltage sensor VS2 are shown. The
first voltage sensor VS1 detects a voltage U12, i.e., a potential difference, between the
first phase line P'l and the second phase line P2. Accordingly, the second voltage sensor
VS2 detects a voltage U23 between the second phase line P2 and the third phase line P3.
To monitor the electric network 1, at least one electric switch element, preferably both
electric switch elements S1_P1, ..., S2_P3, of each phase line P1, P2, P3 is/are opened.
The voltages U12, U23 are then detected and a voltage U13 between the fìrst phase line
P1 and the third phase line P3 is calculated.
14
Furthermore, a speed d (see Figure 2) of the permanent magnet machine M is
determined.
Furthermore, a voltage-dependent variable is determined for each of the voltages U12,
U23, U13, for example, a maximum absolute value during a predetermined time interval of
'100 ms, which follows a determination start time, for example.
A speed-dependent reference variable is determined as a function of the speed d of the
permanent magnet machine M. The reference variable corresponds in this case to the
voltage-dependent variable. For example, the speed-dependent reference variable can be
a minimum absolute value of an idle voltage of the permanent magnet machine M in the
predetermined time intervalof 100 ms.
As explained in greater detail in Figure 2, a difference is then determined between the
speed-dependent reference variable a"nd the voltage-dependent variable and a network
fault is detected if the difference is greater than a predetermined threshold value, in
particular greater than zero.
Figure 2 shows the electric network 1 from Figure 1 and a determination unit BE and an
evaluation unit AE. Reference is made in this case to the statements on Figure 1 with
respect to the design of the electric network 1.
The determination unit BE which can be designed as a microcontroller or parts thereof, for
example, has a signal connection to the voltage sensors VS1, VS2. The voltage U12
between the first phase line P1 and the second phase line P2 and the voltage U23
between the second phase line P2 and the third phase line P3 form input signals for the
determination unit BE. These are filtered by a filter unit F. Furthermore, it is shown that
after the filtering, a difference calculation is performed to determine the voltage U 13
between the first phase line P1 and the third phase line P3. ln the units E, a maximum
absolute value of an amplitude of the voltages U12, U23, U 13 in an interval of 1 00 ms is
determined to determine the voltage-dependent variable. ln an A/D converter AD, a
digitization of this voltage-dependent variable is performed. ln this case, the digitalization
is performed inversely proportional to a level of the voltage-dependent variables. ln
particular, a value of the voltage-dependent variable of zero is coded using the highest
digital value. The digitization can be performed using a Gray code.
15
These digital values, which represent the voltage-dependent variables, form output
signals of the determination unit BE and input signals of the evaluation unit AE, which has
a data connection to the determination unit BE. A further input signal of the evaluation unit
AE is a speed d of the permanent magnet machine M. Via a previously known
relationship, which is shown by way of example in the evaluation block AB in Figure 2, the
evaluation unit AE determines a speed-dependent reference variable Vref (see Figure 3).
Furthermore, it is determined whether the speed-dependent reference variable Vref is
greater by more than a predetermined amount than each of the transmitted voltagedependent
variables. lf the speed-dependent reference variable Vref is greater than at
least one of the transmitted voltage-dependent variables, a fault signal FS is thus
generated. lt is furthermore shown in Figure 2 that it is checked by a further evaluation
block AB2 whether the permanent magnet machine M is not connected to the power
converter C, for example, whether allswitches S1_P1, ..., S2_P3 are open. lt can also be
checked whether no feedback signal, which can also be referred to as a "life signal," from
a traction control unit (not shown) is provided. For example, it can be detected that the
switches S1_P1 , ..., S2_P3 are open if such a feedback signal is not provided.
lf at least one of the conditions is met, a resulting fault signal rFS is thus output by the
evaluation unit AE. A corresponding protective measure can then be initiated as a function
of this resulting fault signal rFS.
Furthermore, a reference signal generation unit RE is shown, which is part of the
evaluation unit AE. lt transmits a reference signal, for example, a step signal, to the digital
converter AD. This can be performed, for example, during startup of the evaluation unit
AE and the determination unit BE. A functionality of the digital converter AD can be
checked in this way.
To check the functionality, it can also be checked whether a bit pattern of the digitized
output signal of the determination unit DE changes by at most 1 bit, in particular if a socalled
Gray code is used for the digitization.
It is not shown that a further electric network can also be monitored by the determination
unit BE and the evaluation unit AE, which also comprises, for example, a power converter
C, phase lines, P1, P2, P3, and a permanent magnet machine M. The determination unit
BE and the evaluation unit AE can thus be used to monitor multiple electric networks,
t6
which each comprise a permanent magnet machine M forthe drive of the railvehicle. lt is
also conceivable that multiple permanent magnet machines of a railroad truck are
monitored.
Figure 3 schematically shows a relationship between a speed d and a speed-dependent
reference variable Vref. The speed-dependent reference variable Vref can be a voltage in
particular. The voltage induced at a specific speed d is approximately linearly dependent
on the motor speed. However, aging influences and temperature, in particular a magnet
temperature, can influence the induced voltage. Thus, for example, the speed-dependent
reference variable Vref can be defined as a minimum speed-dependent voltage, which is
induced at a maximum occurring magnet temperature in normal operation. Alternatively or
cumulatively, the speed-dependent reference variable Vref can also be determined as a
function of the age of the permanent magnet machine, wherein the speed-dependent
reference variable decreases with increasing age. This relationship can be previously
known, however. A reliability and a robustness of the method can advantageously be
increased in this way.
Furthermore, it is shown that as a result of the digitization, the speed d can be assigned to
one speed interval of multiple speed intervals, wherein the speed intervals are delimited
by predetermined speeds d0, d1, d2, d3, d4, d5, d6, d7. An interval-dependent reference
variable VL1, VL2, VL3, VL4, VLs, VL6, VL7 is assigned to each speed interval. For
example, a first interval-dependent reference variable VL1 is assigned to a first speed
interval, which lies between the first speed d0 and a second speed d'1. If the speed is thus
in this speed interval, the speed-dependent reference variable Vref is thus determined as
this interval-dependent reference variable VL1.
t7

We claim:
1 A method for monitoring an electric network (1) in a rail vehicle, wherein the electric
network (1) comprises at least one power converter (C), at least one permanent
magnet machine (M), and at least one first phase line (P1) for the electrical
connection of the at least one power converter (C) and the at least one permanent
magnet machine (M), wherein the first phase line (P1) is interrupted, wherein a
potential difference between a machine-side part of the first phase line (P'1) and a
reference potential and a potential-difference-dependent variable are determined,
wherein a speed (d) of the permanent magnet machine (M) and, as a function of the
speed (d), a speed-dependent reference variable (Vref) are determined, wherein a
deviation of the potential-difference-dependent variable from the speed-dependent
reference variable (Vref) is determined, wherein a network fault is detected if the
deviation is greater than a predetermined threshold value, wherein the electric
network (1) comprises three phase lines (P1, P2, P3), wherein at least two or all
phase lines (P1, P2, P3) are interrupted, wherein a potential difference between a
machine-side part of the phase line (P1, P2, P3) and a reference potential is
determined for each phase line (P1, P2, P3), wherein a speed (d) of the permanent
magnet machine (M) and, as a function of the speed (d), a speed-dependent
reference variable (Vref) are determined for each phase line (P1, P2, p3),
wherein a potential-difference-dependent variable and a deviation of the potentialdifference-
dependent variable from the speed-dependent reference variable (Vref)
are determined for each phase line (P1, P2, P3) and a network fault is detected if at
least one of the deviations is greater than a predetermined threshold value, andior
wherein a minimum potential difference of all potential differences and a variable
dependent on this minimum potential difference are determined, wherein a deviation
of this variable dependent on this minimum potential difference from the speeddependent
reference variable (Vref) is determined and a network fault is detected if
this deviation is greater than a predetermined threshold value, wherein a two-phase
network fault or a three-phase network fault is determined as a function of the
potential differences and/or the at least one deviation.
The method as claimed in claim 1, characterized in that, upon detection of a network
fault, a speed (d) of the permanent magnet machine (M) is reduced and/or a further
interruption of the machine-side part of the phase line (P1, P2, P3) is carried out.
2.
18
3. The method as claimed in any one of the preceding claims, characterized in that an
accuracy of the determination of the potential difference is checked in that at least
one normal operating potential difference is determined in the case of an
uninterrupted phase line (P1, P2, P3), wherein a deviation of the normal operating
potential difference from a normal operating reference value is determined, wherein a
sufficient accuracy is detected if the deviation is less than or equal to a
predetermined threshold value and/or in that a standstill of the permanent magnet
machine (M) is detected, wherein at least one stanQstill potential difference is
detected in the case of an interrupted phase line (P1, P2, P3), wherein a deviation of
the standstill potential difference from a standstill reference value is determined,
wherein a sufficient accuracy is detected if the deviation is less than or equal to a
predetermined threshold value.
The method as claimed in any one of the preceding claims, characterized in that the
speed (d) is assigned to one speed interval of multiple speed intervals, wherein an
interval-dependent reference variable is assigned to one speed interval, wherein the
speed-dependent reference variable (Vref) is determined as the interval-dependent
reference variable.
5. The method as claimed in any one of the preceding claims, characterized in that the
potential difference is assigned to one potential difference interval of multiple
potential difference intervals, wherein an interval-dependent potential difference
value is assigned to one potential difference interval, wherein the potentialdifference-
dependent variable is determined as a function of the interval-dependent
potential difference value.
6. The method as claimed in any one of the preceding claims, characterized in that a
potential difference between a machine-side part of the phase line (P'1, P2, P3) and a
reference potential and a potential-difference-dependent variable is determined for
each phase line (P1, P2, P3), wherein a combined output signal is formed, wherein
the combined output signal comprises a first bit sequence, which codes the potentialdifference-
dependent variable formed from the minimum potential difference, wherein
the combined output signal comprises a further bit sequence, which codes an
equality of all potential differences.
4.
19
7. The method as claimed in any one of the preceding claims, characterized in that the
potential-difference-dependent variable is a maximum amplitude or an RMS value of
the potential difference during a predetermined determination duration.
A device for monitoring an electric network (1)in a rail vehicle, wherein the electric
network (1) comprises at least one power converter (C), at least one permanent
magnet machine (M), and at least one first phase line (P1) for the electrical
connection of the at least one power converter (C) and the at least one permanent
magnet machine (M), wherein the device comprises at least one evaluation unit (AE),
at least one determination unit (BE), and at least one interruption unit, wherein the
first phase line (P1) is interruptible by means of the interruption unit, wherein a
potential difference between a machine-side part of the first phase line (P1) and a
reference potential is determinable by means of the determination unit (BE), wherein
furthermore a potential-difference-dependent variable is determinable, wherein a
speed (d) of the permanent magnet machine (M) and, as a function of the speed (d),
a speed-dependent reference variable (Vref) are determinable, wherein a deviation of
the potential-difference-dependent variable from the speed-dependent reference
variable (Vref) is determinable by means of the evaluation unit (AE), wherein a
network fault is detectable if the deviation is greater than a predetermined threshold
value, wherein the electric network (1) comprises three phase lines (P1, P2, P3),
wherein at least two or all phase lines (P1, P2, P3) are interrupted, wherein a
potential difference between a machine-side part of the phase line (P'1, P2, P3) and a
reference potential is determinable for each phase line (P1, P2, P3), wherein a speed
(d) of the permanent magnet machine (M) and, as a function of the speed (d), a
speed-dependent reference variable (Vref) are determinable for each phase line (P1,
P2, P3),
wherein a potential-difference-dependent variable and a deviation of the potentialdifference-
dependent variable from the speed-dependent reference variable (Vref)
are determinable for each phase line (P1, P2, P3) and a network fault is detectable if
at least one of the deviations is greater than a predetermined threshold value, andior
wherein a minimum potential difference of all potential differences and a variable
dependent on this minimum potential difference are determinable, wherein a
deviation of this variable dependent on this minimum potential difference from the
speed-dependent reference variable (Vref) is determinable and a network fault is
detectable if this deviation is greater than a predetermined threshold value, wherein a
20
two-phase network fault or a three-phase network fault is determinable as a function
of the potential differences and/or the at least one deviation.
L The device as claimed in claim 8, characterized in that the potential difference is
determined close to the motor.
'10. A rail vehicle, comprising a device as claimed in any one of claims B or 9.
Dated this 15th day of March,2017 .
Bombardier Transportation GmbH 2
Þ^/[4ør*,xv
(Dev Robinson)
of Shardul Amarchand Mangaldas & Co.
Attorneys for the Applicant
21
MARKED COPY
We claim:
1. A method for monitoring an electric network (1) in a railvehicle, wherein the electric
network (1) comprises at least one power converter (C), at least one permanent
magnet machine (M), and at least one first phase line (P1) for the electrical
connection of the at least one power converter (C) and the at least one permanent
magnet machine (M), wherein the first phase line (P1) is interrupted, wherein a
potential difference between a machine-side part of the first phase line (P1) and a
reference potential and a potential-d ifference-dependent variable are determ ined,
wherein a speed (d)of the permanent magnet machine (M)and, as afunction of the
speed (d), a speed-dependent reference variable (Vref) are determined, wherein a
deviation of the potential-difference-dependent variable from the speed-dependent
reference variable (Vref) is determined, wherein a network fault is detected if the
deviation is greater than a predetermined threshold value, wherein the elÊçUA
network ('1) comprises three phase lines (P1, P2, P3), wherein at leasttwo erall
phase lines (P1, P2. P3) are interrupted. wherein a potential difference between a
machine-side part of the phase line (P1, P2, P3) and a reference potential is
determined for each phase line (P1, P2, P3). wherein a speed (d) of the permanent
maqnel machine (M) and, as a function of the gpeed (d), a speed-dependent
reference varlable (Vref) are determined for each phase line (P1. P2, P3l
whqrein g_potential-dilference-dependent variaþle, and a deviaiion of the poieniialdifference-
dependent variable from the speed-dependent reference variabie (Vref)
are determined for each phase line (P1, P2, P3) and a network fault is detected if ai
least one of the deviations is greater than a predetermined threshold value, and/or
wherein a minímum potential difference of all potential differences and a variable
dependent on this minimum potential difference are determrned, wherein a deviation
of this variable dependent on thìs minimum potential difference from the speeddependent
reference variable (Vref) is determined and a network fault is detected if
this deviation ìs qreater than a predetermined threshold value, wherein a h¡,¡o-phase
network fault or a three-phase network fault is determined as a function of the
potential differences and/or the at least one deviation.-
L -Êhe+netho4-as-elaimed+n-e{ai+n*1-'Êå€rêe+e+ìzed-irì-fhaï{.he-eleetrie-+e+wer*-(lj
eornpiise+-th+ee'phase-"lines-(P-1-P2, P3Èwhe+eln-atJeast.two-oea{l-phase-{ine+(P,tr-,
P2-P3) are i+terulptedtr+heein-a-peten*al-+i+a.eneeSelween-a-ma€hìr+€-++deçer+,t
MARKED COPY
ef+he-phaseliee-ÉP1, P}P3Ia¡d a refer+nee-pote+lialis-deterpni+ed{o+eaeì+phase
line-(P-l -Pfi P3)t-wher.ein-a-speed-(d)-ef the-pa.rRanent-rRagnel maeh rne {M)-a+lel , -es
a_fu.+et+on-of-{he-speed {d)'-a-speed-dependent r-efereRee-l¡ariable-(Vre{)-are>
deterrnineei for,eaeh phase line (Ptr. P2, P?).
u+hereln-a-po{elltialdjfferenee-e{ependen,t-+ai'iable and a-devìation of Nhe potentialdifferenee*
dependent-varial+le-fror++ the speed-deperrdent referenee-variable (Vre$
are.deterrnined-fo+-eaeh-phas€-tiÊe-(P1 P2-P+þnd-a-net+.¡opk-fault-is-elet€€ted-ì-Ê¿*
least-one-of-the-devìations-is-greater{han-a-p¡,edeternilned-ihrqshold-va}ue-*ndlor
lvherein a-mu+mnum-patent+a-l4,iff.e+enee-e+ati-peteÊ+ialdifferenees and a ',,ariable
dependentor+this minimum potential-differenee-are-deiern+iRedr_whe.rein-a,elevtatjon
et-this+ariabledepende+f-en-Fl+is-minin+u.m-pe[entìe"l-dif{e+enee.{+om-the'+peeddependent
refereRee-veriable-(Vref) is determined-and a.network fardt ie tie,teeted if
t h rs-dev+e*is+ i+ gsa6,1"t-{han-a-pred eterp+rned-{hr"esho ld-ualu e,
3- The-method-as-elairned in"elainr-2'eharaeteri¿ed in fhat-a-fwo-phase-networ-k-fault
hase-neÈwopk-fault is deterrnined as a-f*+etion-sf+he-petent-ia{4ìfferenees
and/or -the at-leasl'one devration
+|-The method as claimed in anyono-ol*he-p+eeed+ngrelaìmsclaim 1,
characterized in that, upon detection of a network fault, a speed (d) of the
permanent magnet machine (M) is reduced and/or a further interruption of the
machine-side part of the phase line (P1, P2, P3) is carried out.
5. 3. The method as claimed in any one of the preceding claims, characterized in
that an accuracy of the determination of the potential difference is checked in that
at least one normal operating potential difference is determined in the case of an
uninterrupted phase line (P1, P2, P3), wherein a deviation of the normal operating
potential difference from a normal operating reference value is determined,
wherein a sufficient accurAcy is detected if the deviation is less than or equal to a
predetermined threshold value and/or in that a standstill of the permanent magnet
machine (M) is detected, wherein at least one standstill potential difference is
detected in the case of an interrupted phase line (P1, P2, P3), wherein a deviation
of the standstill potential difference from a standstill reference value is determined,
wherein a sufficient accuracy is detected if the deviation is less than or equal to a
predetermined threshold value.
1€
MARKED COPY
4. The method as claimed in any one of the preceding claims, characterized in
that the speed (d) is assigned to one speed interval of multiple speed interuals,
wherein an interval-dependent reference variable is assigned to one speed
interval, wherein the speed-dependent reference variable (Vref) is determined as
the interval-dependent reference variable.
5. The method as claimed in any one of the preceding claims, characterized in
that the potential difference is assigned to one potential difference interval of
multiple potential difference intervals, wherein an interval-dependent potential
difference value is assigned to one potential difference interval, wherein the
potential-difference-dependent variable is determined as a function of the intervaldependent
potential difference value.
6 -____The method as claimed in any one of the preceding claims, characterized in
that a potentialdifference between a machine-side part of the phase line (P1, P2,
P3) and a reference potential and a potential-difference-dependent variable is
determined for each phase line (P1, P2, P3), wherein a combined output signal is
formed, wherein the combined output signal comprises a first bit sequence, which
codes the potential-difference-dependent variable formed from the minimum
potential difference, wherein the combined output signal comprises a further bit
sequence, which codes an equality of all potential differences.
7 _The method as claimed in any one of the preceding claims, characterized in
that the potential-difference-dependent variable is a maximum amplitude or an
RMS value of the potential difference during a predetermined determination
duration.
B A device for monitoring an electric network (1) in a railvehicle, wherein the
electric network (1) comprises at least one power converter (C), at least one
permanent magnet machine (M), and at least one first phase line (P1) for the
electrical connection of the at least one power converter (C) and the at least one
permanent magnet machine (M), wherein the device comprises at least one
evaluation unit (AE), at least one determination unit (BE), and at least one
interruption unit, wherein the first phase line (P1) is interruptible by means of the
interruption unit, wherein a potential difference between a machine-side part of the
first phase line (P1) and a reference potential is determinable by means of the
determination unit (BE), wherein furthermore a potential-difference-dependent
MARKED COPY
variable is determinable, wherein a speed (d) of the permanent magnet machine
(M) and, as a function of the speed (d), a speed-dependent reference variable
(Vref) are determinable, wherein a deviation of the potential-difference-dependent
variable from the speed-dependent reference variable (Vref) is determinable by
means of the evaluation unit (AE), wherein a network fault is detectable if the
deviation is greater than a predetermined threshold value-WhçIeiill-r-ç_91_çc,!1,1q
network ('i)comprisesthree phase lines LP1, P2 P_3), wherein at leasttwo orali
phase lines (P1, P2, P3) are interrupted, wherein a potential difference between a
machine-side partof the phase lÌne (P1, P2, P3) and a reference potential is
determìnable for each phase line (P1, P2, P3), wherein a speed (d) of the
permanent magnet machine (M) and, as a function of the speed (d). a speeddependent
reference variable (Vref) are determinedable for each phase_[!q_e]-
P2, P3).
wherein a potential-dÍfference-dependent variabie anlla deviation-qflb-A_p_qJe!!fl_
difference-dependeni variable from the speed-depenclent referencâl¡ariable (Vref)
are determinedable for each phase line (P1, P2, P3) and a network fault iq
detectedable if at least one of the deviations is qreater.than a predetermined
threshold value, and/or wherein a minimum potential difference of all potentíal
differences and a variable dependeni on this mininum potential difference are
deierminedable, wherein a deviation of this variable dependent on thìs minimum
potential differe.nce from the speed-dependent reference variable (Vrefl is
determinedable and a network fault is detectedabie if this deviation is grealer than
a predetermined.threåhold value, wherein a two-phase network fault or a threephase
network fault is determinedable as a function of the potential differences
and/or the at leasi one deviation..
14 9. The device as claimed in claim 148, characterized in that the potential
difference is determined close to the motor.
l¿ 10. A rail vehicle, comprising a device as claimed in any one of claims :l&& or
+9.