The present application is a 371 of International application PCT/DE2012/100009, filed Jan. 16, 2012, which claims priority of DE 10 2011 008 756.7, filed Jan. 17, 2011, the priority of these applications is hereby claimed and these applications are incorporated herein by reference.
The invention relates to a method for determining the position of the rotor of an electrical machine having several phase strands, in relation to the stator, and an apparatus for carrying out the method.
EP 1 005 716 B1 discloses a method for determining the position of the rotor of an electrical machine, wherein the changes of the potential at the star point caused by changes of the potential of the rotor at the current supply inputs are utilized for determining a signal which is representative for the rotor position.
The invention is based on the object of creating a novel method for determining the position of the rotor of an electrical machine which can be carried without picking up an electrical value within the electrical machine.
The method according to the invention which meets this object is characterized in that, for at least one of the current supply inputs of the electrical machine, a change Δİ of the time derivation İ of the current flowing through the respective current supply input is determined as a result of a change ΔU of the potential U on at least one of the current supply inputs of the electrical machine, and a signal representative for the position of the rotor is determined from several changes Δİ occurring simultaneously or successively.
In accordance with the invention, the determination of the position of the rotor takes place exclusively with the aid of currents which can be measured outside of the electrical machine and with changeable voltages.
Preferably, inductivity of one or more phase strands or/and a quotient of these inductivities at a given time are determined as representative measurement signals.
The change ΔU of the potential U takes place within such a short period of time that voltages induced in the phase strands in the meantime and currents flowing through the current supply inputs do not noticeably change.
The phase strands can be switched in a star configuration or/and in a triangular configuration.
Preferably, the change ΔU of the potential U from a potential zero point to a direct voltage takes place, particularly a battery voltage, or vice versa.
In accordance with a particularly preferred embodiment, the change ΔU of the potential U takes place within the scope of a current supply of the electrical machine through pulse width modulation (PWM). The determination of the position now does not require an intervention in the electrical machine itself, or in the manner of operation thereof.
The change Δİ can be determined, for example, by means of a measuring transformer which, for example, produces a voltage signal which may have to be reinforced. For example, in the case of a voltage switch occurring within the scope of the pulse width modulation, a voltage leap results at the secondary winding of the measuring transformer which corresponds to the voltage leap during switching.
Several changes Δİ caused by several potential changes ΔU can be determined for only one of the current supply inputs. The expenses for measurement circuits are correspondingly low. Alternatively, in the extreme case, changes Δİ affected by a single potential change ΔU can be determined for several current supply inputs, so that a measuring circuit is required for each current supply input.
In accordance with a preferred embodiment of the invention, the representative measuring signal is determined from equations which contain the sum of the voltage components over the individual phase strands prior to and after the voltage change.
In the following the invention will be explained in more detail with the aid of embodiments and the enclosed drawing which refers to these embodiments. In the drawing:
Three phase strands 1, 2, 3 of an electrical machine 4 switched in a star configuration each form an inductive resistance 5 and an ohmic resistance 6. The ends of the phase strands 1, 2, 3 facing away from the star point 7 are connected to connections 1′, 2′, 3′ for current supply lines 1″, 2″, 3″.
A current supply device 8 connects the current supply lines 1″, 2″, 3″ through switching devices 9 to 11 with the direct voltage UB of a battery 12 or the voltage zero point corresponding to the pulse width modulation method (PWM-Method).
Measuring devices 13 to 15 in the current supply lines 1″, 2″, 3″ serve for determining changes Δİ, i.e. changes of the first derivation of the current I flowing through the respective current supply line over time. This refers to short changes which result from switching by the switching devices 9 to 11.
The measurement devices 13 to 15 are in connection with an evaluating device 16 which, in turn, is connected to the current supply device 8 through a control line 17.
As illustrated in
With respect to the voltages U1 applied to the individual phase strands İ (İ=1, 2, 3), the following is true at any point in time:
U1=Ui ind+Liİi+RiIi (1),
wherein Ui ind denotes the voltage induced in the phase strand i, Li denotes the inductivity of the phase strand i, and Ri denotes the ohmic resistance thereof.
Depending on whether the battery voltage UB or the voltage zero point contact the phase strand i, Ui=0, Ui=US, Ui=UB−US, or Ui=US−UB, wherein US refers to the potential at the star point 7.
For example, if starting from a switching state in which all three phase strands 1, 2, 3 are connected to the voltage zero point, the following applies:
0=Ui ind+L1İ1+R1I1 (2)
0=U2 ind+L2İ2+R2I2 (3)
0=U3 ind+L3İ3+R3I3 (4).
The ohmic resistances Ri can be considered as being equal and they can be considered to be constant during motor operation of the electrical machine. On the other hand, the inductivities Li depend on the respective influence of the exciter field and, thus, on the position of the rotor relative to the magnetizations of the pole winding cores. Within half a magnetic period there is always an unequivocal relationship between the inductivity of the phase strands and the position of the rotor which can be utilized for determining the position, as will be explained in the following.
After switching, for example, the phase strand 3 to the battery voltage UB by means of the switching device 11, the following result is obtained:
US=U1 ind′+L1İ1′+R1I1′ (5)
US=U2 ind′+L2İ2′+R2I2′ (6)
UB−US=U3 ind′+L3İ3′+R3I3′ (7).
The change of the switching state characterized by the equations (2) to (4) into the switching state according to the equations (5) to (7) takes place so quickly that neither the voltages U1 ind induced in the phase strands nor the currents I change significantly so that Ui ind=Ui ind′ and Ri İ1+R1 I1. The switch essentially only has an effect on ii ie., the first derivation of the currents over time. By subtracting (2)−(5), (3)−(6) and (4)−(7), the following is obtained.
L1(İ1−İ1′)=L1●Δİ1=US (8)
L2(İ2−İ2′)=L2●Δİ2=US (9)
L3(İ3−İ3′)=L3●Δİ3=UB−US (10).
The values Δİ1, Δİ2, Δİ3 can be determined by means of the measuring devices 13 to 15. In the three equations (8) to (10), the inductivities L1, L2 and L3 as well as the potential US at the star point are then unknown.
From the three equations (8) to (10), the ratios L1/L2, L1/L3, L2/L3 can be determined while eliminating US which each represent a dimension for the position of the rotor within half a magnetic period.
In addition, the following applies to the above described switch:
UB=(L3+1/(1/L1+1/L2))●Δİ3 (11).
Consequently, four equations (8) to (14) are available for determining the unknown values L1, L2, L3 and US. Each of the inductivities L1, L2, L3 may serve as a dimension for the position of the rotor within half a magnetic period.
It is understood that for obtaining several equations from which US can be eliminated and L1, L2 or/and L3 can be determined, several successive switching state changes can be considered as long as the condition is adhered to that over the total duration of the measurements the position of the rotor does not perceptibly change. Under these conditions, Δi does not have to be determined in all three phase strands. When considering several switching state changes, the measurement of Δİ in a single phase strand may be sufficient. Accordingly, only a single measuring device is required.
Equations for determining L1, L2, or/and L3 can also be set up for the situation in which the phase strands 1, 2, 3 are switched in a triangular configuration, as illustrated in
A first switching state change concerns, for example, switching of the connection 1a′ from the voltage zero point to battery voltage UB. A second switching state change concerns the switching of the connection 2a′ from the voltage zero point to battery voltage UB. This results in the three equations
Δİ2●L1=UB (12)
Δİ3●L3=UB (13)
Δİ3●L2=UB (14),
from which the unknown values L1, L2, L3 can be determined.
The measuring devices (13) to (15) in the embodiment according to
The above described method for determining the position could be combined with the known method which is based on the evaluation of the potential US at the star point.
The primary winding 20 of the measuring device shown in
From the determined inductivities L1, L2, L3, a flow vector can be determined whose direction coincides with the direction of the rotor flow vector ΦR produced by the magnetic field of the rotor and which is proportional to the rotor flow vector ΦR as long as the currents I1, I2, I3 flowing in the phase strands do not significantly influence the total flow vector Φ. The latter may be the case, especially when the rotor field is weak and the air gap is large. In addition to the rotor flow vector ΦR which is a function of L1, L2, L3, the stator flow vector ΦS, which depends on the currents I1, I2, I3, are determinative for the total flow vector Φ:
Φ(L1,L2,L3)=ΦR+ΦS(I1,I2,I3) (15).
Consequently, in the case of known inductivities L1, L2, L3 and known currents I1, I2, I3, the rotor flow vector ΦR can be computed:
ΦR=Φ(L1,L2,L3)−ΦS(I1,I2,I3) (16).
When the rotor flow vector ΦR is known then the rotary position of the rotor is also known.
The currents I1, I2, I3 can be measured. However, they can also be computed if the inductivities L1, L2, L3 are known, if the values İ1, İ2, İ3 are known (by measuring Δİ1), if the resistances R1, R2, R3 are known, and if the induced voltages U1 ind, U2 ind, U3 ind are known. They can also be computed by means of the above indicated equations.
The voltages Ui ind result from the components Φi of the total flow vector Φ multiplied with the determined rotary speed of the rotor.
Number | Date | Country | Kind |
---|---|---|---|
10 2011 008 756 | Jan 2011 | DE | national |
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/DE2012/100009 | 1/16/2012 | WO | 00 | 7/17/2013 |
Publishing Document | Publishing Date | Country | Kind |
---|---|---|---|
WO2012/097807 | 7/26/2012 | WO | A |
Number | Name | Date | Kind |
---|---|---|---|
5191270 | McCormack | Mar 1993 | A |
5854548 | Taga | Dec 1998 | A |
6366037 | Strothmann | Apr 2002 | B1 |
6771039 | Sakurai et al. | Aug 2004 | B2 |
20050007044 | Qiu et al. | Jan 2005 | A1 |
20090261765 | Stoiber et al. | Oct 2009 | A1 |
20090278485 | Strothmann | Nov 2009 | A1 |
20100262307 | Imura | Oct 2010 | A1 |
20110057593 | Williams et al. | Mar 2011 | A1 |
20110101900 | Basic et al. | May 2011 | A1 |
20110273168 | Strothmann | Nov 2011 | A1 |
Number | Date | Country |
---|---|---|
102006020676 | Nov 2006 | DE |
102007028384 | Dec 2008 | DE |
102008027720 | Dec 2009 | DE |
1005716 | Nov 2001 | EP |
21608312 | Mar 2010 | EP |
2010000640 | Jan 2010 | WO |
Number | Date | Country | |
---|---|---|---|
20150061641 A1 | Mar 2015 | US |