Energy-efficient asynchronous machine

Abstract

In a method for determining a flux of an asynchronous machine, the flux is adjusted according to a loss of the asynchronous machine. Therefore, an apparatus for determining the flux of the asynchronous machine is provided with a model for calculating a loss of the asynchronous machine according to the flux of the asynchronous machine. A selection facility selects a flux according to the loss.

Claims

1. A method for determining a flux of an asynchronous machine, said method comprising: setting the flux as a function of a loss of the asynchronous machine using a non-linear saturation characteristic curve; determining the loss with a model; and either reducing the flux below 100% or increasing the flux above 100%, wherein the model has a thermal model.

2. The method of claim 1, further comprising setting the flux with a minimal loss.

3. The method of claim 1, further comprising optimizing efficiency of the asynchronous machine.

4. The method of claim 1, further comprising setting the flux as a function of load.

5. The method of claim 1, further comprising: calculating a change in the loss as a function of a change in the flux, and setting the flux as a function of the calculated loss.

6. The method of claim 1, further comprising: calculating the loss in a case of a current flux, the loss in a case of a reduced flux, and the loss in a case of an increased flux; determining a minimum of a curve or of a function based on the calculated losses; and setting the flux with minimal loss.

7. A device for determining a flux of an asynchronous machine comprising: a thermal model for calculating a loss of the asynchronous machine as a function of the flux of the asynchronous machine; and a selection facility for selecting the flux as a function of the calculated loss, wherein the device is configured to set the flux as a function of the loss using a non-linear saturation characteristic curve and either reducing the flux below 100% or increasing the flux above 100%.

8. The device of claim 7, further comprising an activation device for said device, said activation device activated as a function of load dynamics of the asynchronous machine.

9. The device of claim 7, wherein the device is integrated in a rectifier.

10. The device of claim 7, wherein the selection facility is configured to optimize efficiency of the asynchronous machine by the selection of the flux.

Description

BRIEF DESCRIPTION OF THE DRAWING

(1) The invention will be explained further by way of example on the basis of exemplary embodiments and with reference to the accompanying drawings, wherein similar elements have the same reference numerals. In the drawings:

(2) FIG. 1 shows a dependence of the efficiency of an asynchronous machine on a specification of the flux;

(3) FIG. 2 shows examples of loss determination;

(4) FIG. 3 shows an asynchronous machine with a rectifier;

(5) FIG. 4 shows measured values for speed, current and motor temperature;

(6) FIG. 5 shows measured values for the flux and a calculated power loss and

(7) FIG. 6 shows a comparison of various optimizations of the asynchronous machine.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

(8) The illustration of FIG. 1 shows a graph which in an asynchronous machine shows by way of example the dependence of the efficiency η 10 on the flux ψ 4. In respect of an axis 9 for the current (I[A.sub.eff]), a curve 11 for the current value I, a curve 12 for the transverse current desired value I.sub.q desired and a curve 13 for the desired value of the d-components of the current I.sub.d desired are also shown. At a value for the flux 4, which is greater than the value for the flux 4 at the maximum of the curve 10 for the efficiency, the intersection of the curve 12 with the curve 13 is approximately (η). At a value for the flux 4, which is smaller than the value for the flux 4 at the maximum of the curve 10 for the efficiency, the minimum of the curve 11 for the current value is approximately (η). This shows that a condition for setting the flux 4 in order to achieve the maximum efficiency (maximum of the curve 10) can still be improved and can be better than the conditions minimum of the current value 11 or intersection of the curves 12 and 13 (I.sub.q desired=I.sub.d desired).

(9) The illustration of FIG. 2 shows examples a) to e) for loss determination in the case of various stationary operating states of an asynchronous machine, A curve 14 with a calculated power loss, in particular the total loss, and a curve 15 with a difference comprising measured active power minus measured mechanical power are shown. The curve 14 has in each case a minimum 16, which is identified by a rhombus. The curve 15 has in each case a minimum 17, which is identified by a square. The dependence on the flux is shown in each case. It can be seen that the minimum of the curve 15 is always in the vicinity of the minimum of the curve 14. For optimization of the asynchronous machine a flux is therefore selected at which the calculated loss (the calculated power loss) is minimal.

(10) The flux value defines the value for I.sub.d desired (i_sd) by way of the saturation characteristic curve. The value I.sub.q desired (i_sq) can be calculated from flux value and torque. The slip results from the rotor resistance, the transverse current and the flux value. The stator frequency also results thereby. All variables for a thermal motor model are known thereby and the motor losses can be unambiguously calculated. The following equations can result thereby:

(11) TABLE-US-00001 i_sq = k_t * Psi * m transverse current i_sd = f(Psi) exciting current (saturation characteristic curve (measured on initial operation) is{circumflex over ( )}2 = i_sd{circumflex over ( )}2 + i_sq{circumflex over ( )}2 stator current f_r = * R_r * i_sq / Psi / 2 / pi rotor frequency f_s = Zp * n + f_r stator frequency

(12) where:

(13) TABLE-US-00002 flux value Psi speed n torque m torque factor K_t stator resistance (thermally adapted) R_s rotor resistance (thermally adapted) R_r number of pole pairs Z_p

(14) The illustration of FIG. 3 shows an asynchronous machine 1, which is fed by a rectifier 3 via a power cable 2. The rectifier 3 has a rectifier regulator 30. A model 5 is integrated therein. This model 5 relates, for example, to a machine model (motor model) and/or a temperature model for the machine. Alternatively or in addition to the temperature model or instead of calculation of the motor losses, the temperatures of the individual motor masses (stator winding, rotor, laminated stator core) can be measured. The temperatures are a type of filtered loss. The colder a motor is at a specified operating state, the more energy-optimized it is. A type of temperature monitoring of the individual motor components with subordinate energy optimization is also conceivable.

(15) The machine can then be optimized by way of an optimization facility 29 in which a selection facility 8 is integrated. For example, a type of optimization can be set and/or an optimum flux set (selected) using the selection facility 8. The optimization relates, in particular, to an optimization of the efficiency. For this, in particular from calculated values for machine losses at magnetic flux values selected so as to be different, a curve is located through these values and the value selected for the flux at which the mathematically generated curve has a minimum. An activation device 31 is provided in order to activate the optimization.

(16) The optimum operating point of the asynchronous machine, or an operating point that comes close to it, is located where the losses calculated by the thermal motor model are smallest. Building on this fact, a search function can be implemented for the flux to be set. In a stationary state of the asynchronous machine, before a real change in the flux, in other words before a change in the desired value for the flux, it can be calculated whether a change in the flux value in an assumed direction (for example increase in the flux or reduction in the flux) would have a positive or negative effect on the efficiency. The flux desired value is then changed accordingly.

(17) The calculation proceeds from the specified load state so that the speed and the torque are defined. Added to this is the assumed flux value. The flux value defines the I_sd by way of the saturation characteristic curve. The I_sq can be calculated from flux value and torque. The slip results from the rotor resistance, the transverse current and the flux value. The stator frequency also results thereby. All variables for the thermal motor model are known thereby and the motor losses can be unambiguously calculated.

(18) The calculated optimum flux is contained for example between 50% and 120%. The calculated optimum flux can be taken into account, for example fully, not at all or in a smoothing manner, in the regulation of the rectifier for the asynchronous machine. Therefore, it is possible, for example, to alternate between a function for efficiency optimization and a loss optimization, with the respective optimization acting fully, not at all or partially, or there being a smooth transition between the optimization variants.

(19) The illustrations of FIGS. 4, 5 and 6 show graphs for measured values in the case of an asynchronous machine. FIG. 4 shows measured values for speed, current and motor temperature. FIG. 5 shows measured values for the flux and a calculated power loss. FIG. 6 shows a comparison of various optimizations of the asynchronous machine.

(20) Thirty load points (six different speeds, five different torques) were approached for recording of the measurements in FIGS. 4 and 5. Three measurements are carried out in all load points:

(21) 1. without flux optimization

(22) 2. with efficiency optimization (I_sd=I_sq)

(23) 3. with loss optimization (minimal loss is optimal).

(24) The temperature of the stator winding was also recorded during the measurement.

(25) In point 2 of the above list the asynchronous machine is optimized in a regulating mode (both servo as well as vector), for example such that the exciting current is set such that its value matches the torque-forming current (I.sub.d=I.sub.q method). This method reduces the flux in particular hi the case of small loads. In the overload range it leaves the flux unchanged. The method I.sub.d=I.sub.q only takes account of the ohmic losses, so that the optimum operating point cannot be achieved.

(26) The illustration of FIG. 4 shows the motor temperature (stator winding temperature) 18, the speed actual value 20 (unsmoothed) and the value of the current actual value 19 (smoothed) over time 21. The following values from left to right were selected as speeds: n=120%, n=100%, n=75%, n=50%, n=30% and n=10%. If the flux value is not optimized (the first measurement in each case), the flux value reaches the expected value of 100% as long as there is no field weakening. The “efficiency optimization” (the second measurement in each case) is effective only due to flux reduction and only in the case of very small loads. In the example the method of loss optimization is nearly always effective: in the case of small loads with flux reduction, in the case of higher loads with flux increase. The total power loss is nearly always reduced in the process. If the asynchronous machine is operated in the range of the nominal point, the optimization potential is reduced if the machine is more or less correctly configured.

(27) In the illustration of FIG. 5 the flux actual value 24 (set flux value) and the calculated total power loss 32 is plotted in addition to the current actual value 19 (smoothed) in the various methods:

(28) 1. without flux optimization

(29) 2. with efficiency optimization (I_sd=I_sq)

(30) 3. with loss optimization (minimal loss is optimal).

(31) As can be seen, in the overload state there is an over-magnetization of the machine as long as the voltage limit is not effective. Furthermore, in certain operating states it can be seen that when the loss optimization is applied (point 3 from above) in the case of flux reduction, more current flows than with the other two methods (point 1 and 2 from above). The total losses are nevertheless reduced. As a result of the flux reduction the reduction in the iron losses is greater than the overall increase in the copper losses in the stator and rotor.

(32) In the illustration of FIG. 6, a distinction is also again made between the three methods:

(33) 1. without flux optimization in a first phase 26

(34) 2. with efficiency optimization (I_sd=I_sq) in a second phase 27

(35) 3. with loss optimization (the minimal loss is optimal) in a third phase 28.

(36) A calculated power loss for the stator 22, a calculated power loss for the stator winding 23, the flux actual value 24 and a calculated power loss for the rotor 25 are also shown.

(37) It can be seen from the graphs in the figures that the efficiency, calculated from the measured variables, is different in the different load states and the different optimizations. Improvements can be achieved in particular at lower speeds or in the partial load range.