Classifications

• • H—ELECTRICITY
  • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
  • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
  • H02P6/00—Arrangements for controlling synchronous motors or other dynamo-electric motors using electronic commutation dependent on the rotor position; Electronic commutators therefor
  • H02P6/14—Electronic commutators
  • H02P6/16—Circuit arrangements for detecting position
• • H—ELECTRICITY
  • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
  • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
  • H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
  • H02P21/22—Current control, e.g. using a current control loop

Definitions

• the present invention relates to rotating electrical machines.
• the present invention more particularly relates to rotating electrical machines provided with a resolver (or resolver), or any other sensor for measuring an absolute angular position, and the present invention relates to the initial setting of this sensor, for example a resolver .
• the shaft of an electric machine When the shaft of an electric machine is equipped with a resolver, it drives the resolver rotor which produces at the output of the windings of its stator a set of alternative electrical signals whose relative amplitude characteristics faithfully reflect and instantly the angular position of the rotor of the machine.
• this signal is used to control the current in the stator windings to maintain an optimum angular difference (typically in quadrature) between the rotating magnetic field in the stator and the magnetic field generated between the poles of the stator. rotor.
• Said rotary electrical machine comprises a stator comprising a stator magnetic circuit which constitutes the active part of the stator. This magnetic circuit is traversed by notches that open into each of its end faces. The notches are filled with conductors that form windings in the magnetic circuit. At the output of the notches in each axial end face of the magnetic circuit, the conductors are folded by forming windings to pass from one slot to the next. Connections windings between them are organized to form induction coils. The ends of the winding conductors which are intended to electrically connect with a suitable connector or junction box.
• the references below are those used in the aforementioned patent application WO 2010/026159.
• This electrical machine comprises a resolver 160 mounted at an axial end of the machine.
• the resolver 160 comprises a resolver stator 164 fixed and axially centered in a housing inside the carcass and a resolver rotor 162 mounted on the shaft 31 of the rotor of the machine, facing the resolving stator.
• the resolver stator 164 is locked in a fixed position axially and angularly in said housing.
• the mounting device comprises a rotational adjustment bearing of the angular position of the resolver rotor 162 on the shaft of the machine rotor 31 and a friction ring 202 for holding the resolver rotor 162 stationary on the shaft 31 of the rotor. the machine in a selected angular position.
• the torque produced depends on the interaction between the rotor flux and the stator flux.
• the rotor flux being produced by the permanent magnets, the torque is regulated by the adjustment of the stator flux for which two parameters are accessible: the amplitude of the flux, itself regulated by the amplitude of the currents of the three-phase system supply, and the phase of the stator flux with respect to the rotor flux.
• This phase is itself regulated by the phase of the stator currents.
• the maximum torque is obtained when the rotor flux is, almost, in quadrature phase with respect to the stator flux.
• the amplitude of the currents is enslaved thanks to regulators which use the measurements of current sensors.
• resolver function is a sensor to measure the absolute position of the resolver rotor on an electric lathe.
• a resolver itself consists of a stator and a rotor. The indication of the measurement depends on the relative position between the resolver stator and the resolver rotor.
• a given reference position between the machine rotor and the machine stator corresponds to a reference value, for example zero, for measuring the rotor position. Since the resolver could be installed in any angle on the machine, there is a difference between the indication of the resolver measurement and the reference value of the rotor position.
• resolver error It is necessary to know this resolver calibration error to know exactly the position of the machine rotor flow and thus to be able to optimize the control of the machine.
• this resolver calibration error there are two solutions to overcome this: either a software compensation or a mechanical adjustment of the relative position between the resolver rotor and the resolver stator so that the indication of the measurement is effectively zero. Call this operation respectively the software setting of the resolver and the mechanical setting of the resolver of the electric machine.
• the angular reference "zero degree" of the absolute position of the rotor of the machine on an electric latch occurs when the rotor poles of the machine (knowing that there may be one or more pairs of poles in the rotor, for example very often three pairs of poles in high performance machines, which then defines three electric revolutions for a mechanical revolution) are aligned on the respective axes of the winding of phase A, c that is to say one of the phases of the stator, whose spatial position is clearly known by the construction of the windings.
• phase A, c that is to say one of the phases of the stator, whose spatial position is clearly known by the construction of the windings.
• This calibration operation has the advantage that one can then, in an industrial object using such an electric machine and manufactured in series as an automobile (for example the electric machine is a motor in an electric power train) , change the electric machine without having to make adjustments to parameters in the control software, for the benefit of the ease of maintenance of these vehicles.
• the object of the invention is to provide useful means to proceed to this calibration of the resolver automatically and accurately.
• the clamping action of the resolver requires two actions: it must be possible to measure the resolver stall error and it is necessary to be able to mechanically adjust the relative position between stator and resolver rotor to reduce this resolver stall error to zero.
• the resolver stall error ie the azimuth deviation of the resolver rotor
• the correction of the position error can be done by blocking the resolver rotor and by giving a train of pulses with the electric machine itself to turn the resolver rotor on the shaft of the electric machine.
• a method for adjusting a resolver of an electric machine comprising a main stator and a main rotor, the main rotor being arranged on a shaft mounted to rotate relative to the main stator, the resolver of said machine comprising a resolver stator and a resolver rotor, the resolver stator being mounted on a support secured to the main stator and the resolver rotor being mounted on a support integral with said shaft, the resolver stator and the resolver rotor being mounted facing one of the another and close to an axial end of said machine, the resolver rotor being the adjustable element of the resolver and being frictionally mounted on its support so that its relative angular position relative to its support can be modified for adjustment by exerting a torque between said resolver rotor and its support, the method comprising the following steps:
• the invention proposes an installation for measuring the stall error of a resolver of an electric machine, the electric machine comprising a main stator and a main rotor, the main rotor being arranged on a shaft rotatably mounted relative to the main stator, the resolver of said machine comprising a resolver stator and a resolver rotor, the resolver stator being mounted on a support secured to the main stator and the the resolver rotor being mounted on a support secured to said shaft, the resolver stator and the resolver rotor being mounted facing each other, the measuring installation comprising a current variator for supplying the main stator, said inverter receiving the angle measurement a m delivered by the resolver, said variator comprising current regulators, preferably of the PI (integral proportional) type, receiving setpoint currents Id and Iq, delivering voltages Uq and Ud supplying a calculation unit performing an inverse Park transformation for applying to the stator wind
• the invention provides a method for measuring the stall error of a resolver of an electric machine, the electric machine comprising a main stator and a main rotor, the main rotor being arranged on a shaft rotatably mounted relative to the main stator, the resolver of said machine comprising a resolver stator and a resolver rotor, the resolver stator being mounted on a support integral with the main stator and the resolver rotor being mounted on a solid support of said shaft, the resolver stator and the resolver rotor being mounted opposite each other, the method using a current controller for supplying the main stator, said variator receiving the angle measurement a m delivered by the resolver and receiving setpoint currents Id and Iq supplying a computing unit performing an inverse Park transformation to apply to the stator windings of the main stator current instructions and appropriate voltage, said inverter comprising a voltage regulator, preferably PI (proportional integral) controlling a voltage Ud and delivering an angle c and correction added
• the invention makes it possible, automatically, to measure the calibration error of the resolver. Note that when this memoir speaks of a resolver, it is necessary to understand any other sensor for measuring an absolute angular position of the rotor of an electric machine.
• FIG. 1 is a simplified diagram of a drive in a calibration bench for implementing the method for measuring the offset of a resolver of a self-driven synchronous machine, according to the invention
• FIG. 2 is a simplified vector diagram relating to an electric machine in no-load rotation (zero current) whose resolver comprises a resolver stall error
• FIG. 3 is a simplified vector diagram relating to the electric machine of FIG. 2, after correction of the error of the resolver
• FIG. 4 schematically represents an adjustment bench for implementing the method for adjusting a resolver of an electric machine according to the invention
• Figure 5 is a block diagram describing the method of adjusting a resolver of an electric machine according to the invention.
• Park transforms is conventional in the art and current controllers for regulating the current (thus the torque) of self-driven synchronous electrical machines built on this basis. are available in the state of the art.
• FIG 1 we see a simplified diagram of an installation for measuring the stall error of a resolver 160 of an electric machine 10. It can be seen that the measuring installation comprises a current variator 5 to power the stator main.
• the drive is designed to exploit Park transformations well known to those skilled in the art.
• the drive receives current instructions Conslq and Consld; it comprises summers 51 receiving at the non-inverting input said current setpoints Cons1q and Cons1d and inverting input values called Mesld and Meslq which will be seen below how they are obtained.
• the variator comprises lines on which the discrepancies between said current setpoints Cons1q and Consld and the so-called Mesld and Meslq values, namely the so-called sld and slq values, circulate. These lines result in the input of current regulators PI (integral proportional) 52 delivering voltages, respectively the voltages Uq and Ud. Note that the means of action of the current regulators PI 52 are the voltages Uq and Ud but these are in fact current regulators which have the effect of respecting said current instructions Conslq and Consld.
• a first computing unit 53 comprising the elements and programs for performing an inverse Park transform on the basis of the Uq and Ud voltages at the output of the PI current regulators 52, and on the basis of the measure the rotor angle cc r obtained as shown below.
• the first computing unit 53 is able to deliver electrical voltage setpoint signals, respectively PWM A , PWM B and PWM C , to be able to generate a system. balanced three-phase AC voltages.
• a power stage 54 receives the PWM instructions A , PWM B and PWM C from the first calculation unit 53; it also receives a power line on which electrical energy is available in the form of DC-DC direct voltage.
• the power stage 54 is able to generate a balanced three-phase voltage system, respectively U A , U B and Uc, for supplying each of the phases A, B and C of the electrical machine 10.
• the current controller 5 also comprises a signal processing unit 56 for transforming the electrical signals received from the resolver stator 160 to measure angle ⁇ m .
• the current controller 5 comprises a second computing unit 58 comprising elements and programs for performing a direct Park transformation. From the currents measured on each of the phases, respectively the measurements Mes I A , Mes 1 ⁇ 2 and Mes le and the rotor angle a r , as known per se by those skilled in the art, the second calculation unit 57 delivers the said Mesld and Meslq values.
• the current converter 5 further comprises a PI 60 voltage regulator, PI type (integral proportional), delivering an angle a c for correcting the calibration error by regulating a setpoint value for the voltage Ud
• PI 60 voltage regulator
• PI type integrated proportional
• the voltage Ud is fed to the inverting input of an adder 59, which receives as a non-inverting input a setpoint ConsUd.
• the voltage difference sUd supplies the voltage regulator PI 60.
• Figure 2 This is a vector diagram, resulting from a direct Clark transformation, of an electrical machine operating at zero (zero current) and whose resolver has a wedge angle error. c o.
• the machine rotates at such a speed that it generates an electromotive force E. Therefore, to control a zero current, the current converter 5, in which the voltage regulator PI 60 is deactivated, generates a voltage U exactly identical in phase. and amplitude to the force
• This voltage U breaks down into voltages Ud and Uq on the two respective axes d and q of the Park mark. If there is a calibration error, ie aco not equal to zero, the component Ud is present. If there is no calibration error, either CCcO zero, the component ⁇ Ud is zero, the component Uq is equal to the voltage 3 ⁇ 4 itself equal to the electromotive force E, all these voltages being in phase with the axis q of the Park mark: this is the case of Figure 3.
• the procedure is as follows.
• the main rotor of the electric machine which is mechanically free, is rotated by injecting appropriate currents through the variator (described above and illustrated in FIG. 1) until a certain speed of rotation is achieved.
• this speed must be large enough for the Ud and Uq voltage determination dispersions by the PI current regulators 52 to be sufficiently small compared with the amplitude of the electromotive force E created by the rotation.
• the current setpoints - Conslq and Consld - are set to 0. This results in a deceleration of the rotor of the electric machine, "coasting".
• the output of the current regulators PI 52 (in the drawing, the current regulators PI 52 for the current Iq and respectively for the current Id) which are respectively the voltages Uq and Ud will take the necessary value to exactly compensate the electromotive force of the electric machine and thus effectively cancel the current in the electric machine.
• the Ud component should be zero.
• the installation allows, thanks to the voltage regulator PI 60, to perform a software correction of the angle given by the resolver measurement to allow the drive to operate with the right reference rotor position measurement. Under these conditions where the drive would operate with the correct rotor position measurement reference, there would be a zero Ud voltage.
• the regulation of the voltage Ud is controlled so as to reach zero while the speed of rotation of the main rotor of the machine is still greater than a predetermined threshold. It is the accuracy of the measurement because the voltage on the axis d, for a stall error, will be even greater than there is electromotive force at the terminals of the electric machine.
• the steps consisting of (i) applying to the main stator a current setpoint until the machine is repeated are repeated.
• a first rough estimate of the calibration error obtained for example by injecting constant currents in two phases of the main stator to obtain an equilibrium position of the resolver rotor and compare this equilibrium position to the theoretical alignment position for this current injection.
• This coarse calibration allows us to apply current setpoints, respectively Conslq and Consld, capable of providing sufficient torque to start the motor at a sufficient speed and in the right direction of rotation, a prerequisite for applying the method of measure of the angle a c o, more precise and more direct, which has been explained previously.
• FIG. 4 shows a bench 1 for correcting the stall error of a resolver of an electric machine.
• Each pulse of the pulse train of step (c) above results in rotating the resolver rotor by a predetermined angle.
• the pulses and the predetermined angle can be determined experimentally.
• the electric machine develops a motor rotation torque but, as the resolver rotor is mechanically locked relative to the resolver stator, and as the magnitude of the torque exceeds the torque due to the friction of the resolver rotor 162 on its support, that is to say on the bearing surface of the shaft 31 intended to receive said resolver rotor 162, a sliding is effected, that is to say a forced rotation of the resolver rotor 162 by This is how the adjustment of the resolver rotor 162.
• the number of pulses is calculated by dividing the position angle error by said predetermined angle. Preferably, said predetermined angle is reduced at each iteration of the adjustment process.
• the bank 1 receives and immobilizes (by means not shown) the synchronous electric machine 10 to correct.
• This machine comprises a main stator (not shown) and a main rotor (not shown) rotating relative to the main stator and mounted on a shaft 31 which shows a splined end intended to be coupled with a mechanism driven by or driving the electric machine.
• This machine comprises a resolver stator 164 and a resolver rotor 162.
• the end of the shaft 31 opposite the corrugated end has a shoulder against which the resolver rotor 162 engages.
• a friction ring 202 which, while allowing a relative rotation slip if the applied torque is sufficiently high, immobilizes the resolver rotor 162 on the shaft 31 to the service requests, ie in normal operation. It is one technological means among others to immobilize a resolver rotor on the shaft of the electric machine to service stresses while retaining an ability to adjust the azimuth resolver rotor.
• the end face of the resolver rotor is provided with two blind holes 209 and 210.
• the bench 1 for correcting the stall error also comprises a locking member 4 which comprises a head 41 having two studs 42 shaped and positioned relative to each other to be inserted into the holes blind pins 209 and 210.
• the pins 42 are mounted on the head 41 in an elastic manner, allowing them to sink somewhat into it if we press their end by a parallel movement to the axis of the rotor 31.
• the head 41 is mounted on an actuator 40 centered on the axis of the rotor 41 and capable of moving the head 41 in the direction of the axis of the rotor 31 to bring the head 41 closer to or away from the resolver rotor 162, and able to rotate around the axis of the rotor 31 the head 41 so that, by combining a rotational movement of the head 41 and a translational movement thereof, the pins 42 can engage with the blind holes 209 and 210.
• the locking member 4 finally has a brake 43 allowing immobilize the actuator 40 and thus the head 41.
• FIG. 5 is a block diagram describing the method of adjusting a resolver of an electric machine according to the invention.
• a bank 1 for correcting the calibration error according to the invention preferably comprises a controller for automating operations.
• Such an automaton contains in memory the parameters making it possible to treat a whole range of electrical machines.
• the operator selects the type of machine installed on the bench: we see a first block 61 of "program selection" selecting a program adapted to the electric machine to be adjusted. This program automatically links the operations described below.
• a block 62 of "measuring offset a c o" making a measurement preferably by means of a measurement system comprises a current controller 5 and a voltage regulator PI 60 delivering an angle a c o measurement of the calibration error, as explained above.
• a test block 63 checking if a c o (the resolver rotor position error) is less than a pre-established tolerance, here equal to 0.1 °. If this is the case, the resolver setting is complete (exit to the "end" status). If this is not the case, the program goes through block 64 "blocking rotation of the resolver": referring to FIG.
• this action consists in first controlling a fast approach of the head 41 towards the resolver rotor 162, then a slow approach on the last millimeters. Once in contact, it controls a rotation of the head 41 until the pins 42, by their elastic mounting on the head 41, are inserted spontaneously into the blind holes 209 and 210.
• the resolver rotor 162 is then secured to the head 41, itself secured to the actuator.
• the actuator is then immobilized by the brake 43, which immobilizes the head 41 and therefore also the resolver rotor 162.
• this operation is to supply the main stator of the electric machine to perform torque pulses, for a given time, by appropriately controlling the drive of the machine. electric, as explained above. Then it is linked by the block 66 of "resolver rotor release” which consists of releasing the brake 43 and rolling back the head 41. A new measurement is then made, by returning to the block 62 of "offset measurement a c o", until the resolver rotor is seated within the agreed tolerance.
• the final assembly of the electrical machine can generally be carried out by mounting a protective cover such as that described for example in the aforementioned WO 2010/026159 patent. .

Applications Claiming Priority (2)

Publications (1)

Family

ID=43603055

Family Applications (1)

Country Status (6)

Families Citing this family (27)

Citations (1)

Family Cites Families (18)

• 2010
  • 2010-05-21 FR FR1053962A patent/FR2960358B1/fr not_active Expired - Fee Related
• 2011
  • 2011-05-05 EP EP11718088A patent/EP2572447A1/de not_active Withdrawn
  • 2011-05-05 WO PCT/EP2011/057228 patent/WO2011144457A1/fr active Application Filing
  • 2011-05-05 JP JP2013511596A patent/JP5847166B2/ja not_active Expired - Fee Related
  • 2011-05-05 CN CN201180024998.3A patent/CN102906989B/zh not_active Expired - Fee Related
  • 2011-05-05 US US13/698,842 patent/US8933655B2/en not_active Expired - Fee Related

Patent Citations (1)

Also Published As

Similar Documents

Legal Events