Method and apparatus for determining initial rotor position of a three-phase motor

Abstract

The present application provides a method for determining an initial rotor position of a three-phase motor (310) using a control circuit (3100), comprising energizing a first phase (312.1) and a second phase (312.2) of the three-phase motor (310) with a current that increases or decreases during a first energization interval; determining an induced voltage in a non-energized third phase (312.3) of the three-phase motor (310) at at least two measurement time points separated in time during the first energization interval; determining an induced voltage difference (3700) of a voltage induced in the third phase (312.3) between the two induced measurement time points separated in time by taking into account a voltage change caused in the control circuit (3100) of the first phase (312.1) and / or the second phase (312.2) by energizing the first phase and the second phase in the first energization interval in the form of a predetermined, estimated shunt voltage difference; and determining the initial rotor position based on the determined induced voltage difference (3700).

Description

[0001] This application is a divisional application of application number 202280025819.6, application date March 25, 2022, and invention name “Method and device for determining magnetization parameters of a three-phase motor”. Technical Field

[0002] The present invention relates to a method for providing commutation intervals for a three-phase motor, a control unit for a three-phase motor, and a three-phase motor. The present invention also relates to a method and apparatus for determining magnetization parameters of a three-phase motor. The present invention also relates to a method, a control unit, and a motor for determining an initial rotor position of a three-phase motor. Therefore, the present invention particularly relates to the field of motors and motor control. Background Art

[0003] For reliable operation of three-phase electric motors, it is often advantageous to detect the rotor position, especially when the motor is at a standstill or at low speed. In particular, knowledge of the rotor position facilitates accurate and reliable commutation. For this purpose, various methods are known in the prior art, which can be divided into sensor-based methods and sensorless methods. A disadvantage of sensor-based methods is that suitable sensors are required and the hardware effort and associated costs are higher than for sensorless methods. Sensorless methods are usually based on feeding test pulses or measurement pulses into the motor, which can result in undesirable noise. Sensorless methods are usually based on the position and current dependence of the stator inductance of the motor. Various types of sensorless methods are described, for example, in the published patent application DE 10 2019127 051 A1.

[0004] The application limits of sensorless methods for detecting rotor position are still the subject of research and development and can often only be determined through extensive experimentation. This can also lead to misunderstandings that may only become apparent late in the product development process, leading to corresponding problems and / or requiring considerable effort and high costs to eliminate.

[0005] While it's convenient in principle to simulate and predict the characteristics of a motor or application, this has traditionally been impossible because it requires precise knowledge of the motor's magnetization parameters. These parameters include k1, which represents the change in inductance due to the rotor magnets, and k2, which represents the change in inductance due to the current supplied to the motor.

[0006] Traditionally, the determination of magnetization parameters, in particular k2, has only been possible empirically. Simple setups for direct determination by measurement have not been successful to date, particularly due to the influence of parasitic resistances in the overall setup. In principle, it is possible to determine the inductance L of a single winding of the motor by measuring the inductance L of the single winding of the motor. D and L QTo directly determine the parameter k1, however, direct measurement of the individual windings of the motor is not possible in most cases. This is because only the three outer terminals of a three-phase motor are usually accessible. Internally, the motor is connected in star or delta. This results in overlapping influences on at least two of the three windings, making a direct calculation no longer possible.

[0007] Sensorless methods for determining the initial rotor position are typically based on measuring the voltage at an inductive voltage divider, i.e., the voltage at the junction of the de-energized phase. However, a disadvantage of these methods is that reliable north / south detection is not possible in some cases. This is particularly true for high-current motors.

[0008] In order to achieve reliable switching between commutation intervals, DE 10 2019 127 051 A1 proposes a method in which bipolar pulse width modulation is applied to the connections of two of the three phases and switching to the next commutation interval occurs when a predetermined voltage threshold is reached.

[0009] DE 10 2016 123707 A1 describes a control device for an electric motor. Here, the voltage between the star point and a reference potential can be tapped using passively switched phase connections. Care must be taken to avoid current flowing through the passive phases, which could materially affect the voltage divider formed at the star point.

[0010] WO 2009 / 047217 A2 describes a method for operating a three-phase electric motor. In this method, the deviation of the time curve of the induced voltage compared to the time curve of the pulse voltage is determined to determine the rotor position.

[0011] DE 10 2016 123715 A1 describes a control device for a multi-phase electric motor having a control device. The control device is configured to stamp a pulse-width-modulated voltage pattern into four phases of a five-phase connection, so that an evaluation signal that depends on the rotational angle of the multi-phase electric motor is generated at a first phase connection. The control device is further configured to determine the rotational angle and / or commutation condition of the multi-phase electric motor based on the evaluation signal.

[0012] DE 102 21 385 A1 describes a method for starting a brushless DC motor, wherein a measuring shunt and a comparator are used to compare the phase currents in the winding phases.

[0013] DE 102 20 077 A1 describes a method for starting a brushless DC motor. Summary of the Invention

[0014] The object of the present invention is to provide a method and a control unit for a three-phase electric motor, with which the reliability of the commutation can be increased and the hardware requirements can be kept low.

[0015] In a first aspect, the present invention relates to a method for providing a commutation interval for a three-phase motor. The method comprises providing an initial commutation interval for operation of the three-phase motor, wherein the commutation interval has an upper switching limit g o,0 and switching lower limit g u,0 , and determining the operating current (i) occurring during operation of the three-phase motor. Furthermore, the method comprises adapting the initial commutation interval according to the determined operating current (i), wherein the switching upper limit g of the commutation interval is performed o,i With the switching lower limit g u,i The distance between them is adapted proportionally to the square of the operating current (i), and / or the offset of the commutation interval is adapted linearly proportionally to the operating current.

[0016] In another aspect, the present invention relates to a control unit for a three-phase motor. The control unit is configured to provide an initial commutation interval for operation of the three-phase motor, wherein the commutation interval has an upper switching limit g o,0 and switching lower limit g u,0 Furthermore, the control unit is configured to determine an operating current i occurring during operation of the three-phase motor or intended or used for operation of the three-phase motor. Furthermore, the control unit is configured to perform an initial adaptation of the commutation intervals as a function of the determined operating current i, such that a switching upper limit g of the commutation intervals is achieved. o,i With the switching lower limit g u,i The distance between them is adapted in proportion to the square of the operating current i and / or the offset of the commutation interval is adapted in linear proportion to the operating current.

[0017] In another aspect, the invention relates to a three-phase electric motor comprising a control unit according to the invention.

[0018] A three-phase motor is an electric motor with a stator and a rotor that has three phases or three windings, each of which can be energized via its own connection. The windings can be arranged in the motor so that they at least partially overlap. For example, the phases of the motor can be connected to one another in a delta connection or in a star connection. The terms three-phase motor, electric motor, and motor are used synonymously in this disclosure.

[0019] Here, a commutation interval is a range of values ​​for the back-measured voltage on the de-energized phase within which the predetermined commutation remains unchanged. A commutation interval can correspond to a specific range of rotation angles of the rotor relative to the stator of a three-phase motor, but these are not necessarily directly comparable. Within the corresponding commutation interval, the commutation can be selected so that a motor with the predetermined commutation produces the maximum effect and / or torque compared to other commutation options. If the voltage on the de-energized phase leaves the corresponding commutation interval, a change in the commutation interval may be necessary or advantageous to achieve the maximum possible torque and / or effect. A change in commutation here means a change in the phase assignment and / or the direction of the current in the phases, i.e., which two phases are energized or which phase remains de-energized. Depending on the direction of rotation of the rotor relative to the stator, the change between commutation intervals or between energized phases can be in opposite directions. In the context of this disclosure, the term commutation interval is intended to refer to a voltage range. An initial commutation interval corresponds to an initial starting value of the commutation interval that has not yet been adapted to the operating current by the method according to the present invention or is subject to further adaptation.

[0020] Switch upper and lower g limits o,0 and g u,0 Here, is the voltage value representing the end point of the (initial) commutation interval. When the voltage corresponding to the upper switching limit is reached and / or exceeded, a change to the subsequent commutation interval may be necessary or advantageous. When the voltage corresponding to the lower switching limit is reached and / or fallen below, a change to the previous commutation interval may be necessary or advantageous.

[0021] Switch upper limit g o,i and switching lower limit g u,i The distance between them determines the width or size of the commutation interval. A commutation interval shift can mean a shift of the center of the commutation interval along the rotation angle of the rotor or the associated voltage, without, however, an accompanying change in the width or size of the commutation interval.

[0022] The operating current is the current or current intensity with which a three-phase motor energizes the energized phases during operation. The operating current can depend, in particular, on the load and / or load on the three-phase motor. The operating current can also be referred to as the load current. In particular, the operating current can influence the voltage on the de-energized phases and lead to offsets in the determination of the rotor rotation angle and / or commutation interval.

[0023] The present invention offers the advantage that, when providing the commutation intervals, the operating current can be taken into account as an influencing variable, and corresponding undesirable deviations of the commutation intervals from the ideal commutation intervals can be reduced or even completely avoided. This offers the advantage that the reliability of the control of the three-phase motor can be increased. It also offers the advantage that a three-phase motor controlled with the commutation intervals provided according to the invention can be operated with high efficiency, since high torque can be reliably achieved and internal energy losses can be reduced. Furthermore, the present invention offers the advantage that the noise generated during operation of the correspondingly controlled three-phase motor can be reduced by the commutation intervals provided according to the invention.

[0024] Optionally, the switching upper limit g o,i and the switching lower limit g u,i Predefined functional parameters are provided for adapting the distance between the switching limits proportional to the square of the operating current, and / or for adapting the offset linearly proportional to the operating current. Optionally, adapting the commutation interval further comprises calculating the adapted distance and / or adapted offset using the predetermined functional parameter and the current operating current. This offers the advantage that the distance between the switching limits and the offset of the commutation interval can be adapted using a simple algorithm based on the predetermined functional parameters, and accordingly, the hardware and / or computational effort required to provide the adapted commutation interval can be kept low.

[0025] Optionally, a plurality of adapted commutation intervals are provided for different operating currents (i), and optionally, adapting the commutation intervals includes selecting one of the provided adapted commutation intervals based on the current operating current. In other words, several predetermined commutation intervals are provided, from which one is then selected based on the current operating current. For example, one or more value ranges may be assigned to each provided adapted commutation interval. If the determined operating current is within one or more assigned value ranges, the corresponding adapted commutation interval may be used to operate the motor. If the operating current changes and assumes a value outside the previously assigned value range, a different commutation interval may be selected accordingly and used for further operation of the motor.

[0026] Optionally, the method includes providing predetermined threshold values, wherein the commutation interval is adapted when an operating current (i) corresponding to one of the threshold values ​​is present. For example, the determined operating current can be compared with predetermined threshold values, and when the corresponding threshold value is reached, exceeded, or fallen below, the commutation interval associated with the corresponding threshold value can be used. This has the advantage of being easy to implement and requiring minimal data storage. The data can be stored, for example, within the control unit for control purposes and / or externally transmitted to the control unit.

[0027] Optionally, the method further comprises determining additional threshold values ​​by interpolating and / or extrapolating the predetermined threshold values, wherein the commutation interval is adapted when an operating current corresponding to a specific one of the additional threshold values ​​is present. In other words, the method may comprise interpolating and / or extrapolating to generate further additional threshold values ​​based on the predetermined threshold values, which may then be used to select the commutation interval as a function of the operating current. This enables a more refined coordination and allocation of the commutation interval to the determined operating current.

[0028] Optionally, the switching upper limit g of the commutation interval is determined according to the opening of the parabola and according to the displacement of the optional vertex. o,i and switching lower limit g u,i The distance between them is adapted in proportion to the square of the operating current. In other words, the adaptation of the distance between the switching limits is based on a parabolic function, which is characterized by the opening of the parabola and the y-intercept. The parabola here corresponds to the voltage difference between the upper and lower switching limits (vertical axis) and the operating current (horizontal axis). For the mean-free measurement pulse generation method, the parabola is axisymmetric about i0=0. The vertical position of the vertex here represents the voltage difference that the upper and lower switching limits of the commutation interval should have if the initial commutation interval is not adapted according to the operating current. However, when using the mean-value measurement pulse generation method, the symmetry axis can also be offset and be at the value i0≠0. In this case, the vertex of the parabola is not on the vertical axis i=0, but is offset by the value i0≠0. This enables simple parameterization to adapt the distance between the upper and lower switching limits of the commutation interval.

[0029] Optionally, the offset of the commutation interval can be adapted linearly proportional to the operating current based on the slope of a straight line and an optional zero offset. This straight line describes the linear dependence of the offset of the mean value of the commutation interval, which represents the voltage value by which the commutation interval is offset compared to the initial commutation interval as a function of the operating current. In the method for generating measuring pulses without a mean value, the straight line passes through the coordinate origin. In the method for generating measuring pulses using a mean value, the zero offset must also be taken into account. This allows for simple parameterization for adapting the offset between the upper and lower switching limits of the commutation interval.

[0030] Optionally, the switching upper limit g of the initial commutation interval is o,0 and switching lower limit g u,0 The predetermined voltage values ​​correspond to or are based on the voltage across the inductive voltage divider of the three-phase motor. o,0 and g u,0 (initial) or g o,i and g u,i(Adaptation to the operating current) Alternatively, the following form can be employed: different voltage values ​​and voltage differences at different measurement times are recorded. A first voltage value can be generated by feeding a current pulse with a voltage pulse of a first polarity, and a second voltage value can be generated by feeding a current pulse with a voltage pulse of a second polarity opposite to the first polarity. This offers the advantage that the reliability of determining the respective switching limit can be increased, since interference influences can be reduced by using the voltage difference.

[0031] Optionally, the method further comprises adapting the initial commutation interval according to a determined voltage value of a supply voltage supplied to the three-phase motor. This offers the advantage that any influence of the supply voltage on the commutation interval can also be taken into account, and any resulting deviation of the commutation interval from an optimal commutation interval can be reduced or avoided.

[0032] Optionally, adapting the initial commutation interval according to the determined voltage value of the supply voltage comprises normalizing the voltage at the inductive voltage divider to the determined voltage value of the supply voltage. The voltage divider can optionally be formed by a connection of a de-energized phase of the electric motor, wherein the voltage divider is caused by the different inductances of the first and second energized phases of the electric motor due to the rotational angle. The voltage at the inductive voltage divider serves as a measured value for comparison with an upper and / or lower switching limit of the commutation interval, wherein a commutation change occurs when the voltage at the inductive voltage divider reaches one of the switching limits. By normalizing the voltage at the inductive voltage divider with the determined voltage value of the supply voltage, any influences that lead to an undesirable deviation of the voltage at the voltage divider compared to the switching limits of the commutation interval can be reduced or avoided.

[0033] According to an alternative embodiment, the offset of the commutation intervals may be related to the offset of the commutation intervals of the energized motor and satisfy the following mathematical relationship:

[0034] (1)

[0035] Here, the subscripts o and u denote the upper and lower switching limits, the subscript i denotes the assignment of the switching limits to the motor operation with a certain operating current (i), i.e. the load current to which the respective switching limits are adapted, and the subscript 0 denotes the corresponding initial switching limit for the de-energized motor.

[0036] The distance between the upper switching limit go,i and the lower switching limit gu,i may optionally satisfy the following mathematical law:

[0037] (2) A = g o，i -g u，i

[0038] A further object of the present invention is to provide a method and a device by means of which the magnetization parameters k1 and k2 can be determined reliably and optionally automatically.

[0039] Another aspect of the present invention therefore also relates to a method for determining magnetization parameters of a three-phase electric motor having a rotor and a stator. The method comprises feeding a current pulse into a first phase and a second phase of the three-phase electric motor, wherein the current pulse is fed during a measurement period, and rotating the rotor relative to the stator by at least a portion of an electrical revolution during the measurement period, the rotation occurring uniformly throughout the measurement period. The method further comprises measuring the voltage across an inductive voltage divider of the three-phase electric motor during the measurement period and determining a measured value of the voltage measured at a predetermined rotation angle of the rotor relative to the stator. Furthermore, the method comprises calculating a corresponding simulated value of the voltage at the voltage divider at a predetermined rotation angle using a predetermined estimated value of the magnetization parameter and adapting the predetermined estimated value of the magnetization parameter such that a deviation between the simulated value of the voltage and the measured value of the measured voltage is minimized. Furthermore, the method comprises determining at least one or more magnetization parameters of the three-phase electric motor based on the adapted estimated value of the magnetization parameter in such a way as to minimize a deviation between the simulated value and the measured value of the measured voltage (U3).

[0040] In another aspect, the present invention relates to a device for determining magnetization parameters of a three-phase motor having a rotor and a stator. The device comprises a control unit for energizing the three-phase motor and for feeding current pulses to a first phase and a second phase of the three-phase motor during a measurement period. The device further comprises a rotation unit for rotating the rotor relative to the stator, wherein the device is configured to uniformly rotate the rotor relative to the stator by one electrical revolution via the rotation unit throughout the entire measurement period; and a measuring element for measuring a voltage at an inductive voltage divider of the three-phase motor during the measurement period. The device further comprises a control unit configured to determine a measured value of a voltage measured by the measuring element at a predetermined rotation angle of the rotor relative to the stator, and to calculate an analog value of the voltage across the voltage divider at the predetermined rotation angle using a predetermined estimated value of the magnetization parameter and to adapt the predetermined estimated value of the magnetization parameter so that a deviation between the analog value of the voltage and the measured value of the measured voltage is minimized. The control unit is further configured to determine at least one or more magnetization parameters of the three-phase motor based on the adapted estimated value of the magnetization parameter in such a way as to minimize a deviation between the analog value and the measured value of the measured voltage (U3).

[0041] A three-phase motor is an electric motor with a stator and a rotor that has three phases or three windings, each of which can be energized via its own connection. The windings can be arranged in the motor so that they at least partially overlap. For example, the phases of the motor can be connected to one another in a delta connection or in a star connection. The terms three-phase motor, electric motor, and motor are used synonymously in this disclosure.

[0042] In particular, current pulses can be fed in by applying voltage pulses to the corresponding phase connections. The direction of the current is influenced by the polarity of the applied voltage. However, due in particular to the inductance of the windings or phases, the direction of the current does not always necessarily follow the polarity of the voltage pulse. For example, a rapid reversal of the polarity of the applied voltage pulse can result in a reduction in the previously existing current without reversing the direction of the current.

[0043] The inductive voltage divider can correspond in particular to the voltage between the connection of the de-energized phase and ground. This voltage is determined by the ratio of the inductances L1 and L2 of the first and second energized phases, which in turn depends on the position of the rotor.

[0044] The fact that the voltage measurement is performed at a predetermined rotation angle means that the measured voltage value is measured at least at a fixed rotation angle of the rotor. Here, the predetermined rotation angle can refer to a relative angle relative to the initial rotation angle position of the rotor. For example, a specific measurement interval can be spanned by a predetermined rotation angle. Alternatively or additionally, the predetermined rotation angle can depend on the measurement schedule and correspond to an angle at which the measured voltage has a minimum, maximum, or another identifiable value.

[0045] Calculating the analog value of the voltage in a manner that minimizes the deviation of the analog value of the voltage from the measured value of the measured voltage means that the analog value is optimized so that the analog value of the voltage is as close as possible to the measured voltage value. In other words, the analog value is optimized by varying the parameters k1 and k2.

[0046] The present invention offers the advantage of being able to reliably determine the magnetization parameters k1 and k2. For example, the magnetization parameters for each motor type and / or each individual motor can be determined at the factory, and the parameters determined in this way can be used to reliably determine the rotor position during operation of the motor. The present invention thus offers the advantage that the magnetization parameters determined according to the invention enable a sensorless determination method for the rotor position, which would not be applicable if the magnetization parameters were not reliably determined. The present invention thus offers the advantage that, even for motors requiring reliable knowledge of the rotor angle, there is no need to resort to sensor-based determination of the rotor angle, which carries with it additional costs.

[0047] Furthermore, the present invention offers the advantage that the determination of the magnetization parameters can be partially or fully automated. This enables the magnetization parameters of numerous motors to be routinely determined. For example, the magnetization parameters of a large number of motors can be determined in this way, and the averaged magnetization parameters can be used for the corresponding series of motors.

[0048] Optionally, measuring the voltage at the inductive voltage divider includes measuring a voltage difference between a first voltage value and a second voltage value. The first voltage value is generated by feeding a current pulse with a voltage pulse of a first polarity, and the second voltage value is generated by feeding a current pulse with a voltage pulse of a second polarity opposite to the first polarity. This offers the advantage of increasing the reliability of the determination because interference effects can be reduced by using the voltage difference. Specifically, such interference effects that affect both the first and second voltage values ​​or the current pulses generated thereby can be averaged so that they have no effect on the voltage difference used to determine the magnetization parameter. In particular, using the voltage difference can reduce motor voltages caused by the rotation of the rotor relative to the stator, which would otherwise be superimposed on the measured voltage.

[0049] Optionally, the current pulses are designed as test pulses, wherein, in addition to the test pulses, the method is performed at least once with the three-phase motor de-energized. Determining one or more magnetization parameters includes determining a magnetization parameter k1, which characterizes a change in inductance caused by one or more rotor magnets of the three-phase motor. Performing the method with the motor de-energized allows the influence of the rotor magnets on the inductance of the motor, characterized by the parameter k1, to be determined.

[0050] Optionally, in addition to the test pulses, the method is performed at least once with the three-phase motor energized, and determining the one or more magnetization parameters includes determining a magnetization parameter k2 that characterizes the change in inductance caused by energizing the three-phase motor. This offers the advantage that the effect of energizing the motor on the inductance can also be reliably and easily determined. Optionally, the three-phase motor is energized such that the commutation direction of the energization remains unchanged during the measurement period. In other words, the rotor's rotational angle changes while the commutation direction of the energized motor remains unchanged. In this way, the effect of the rotor's rotational angle on the change in inductance caused by energization can be reliably determined.

[0051] Optionally, the method is performed at least once while the three-phase motor is powered, wherein the current pulses are designed as block commutation pulses. This offers the advantage that, when the motor is powered, no additional measurement pulses need to be provided for the method to be performed, but rather conventional commutation pulses can be used. This reduces the measurement effort and / or complexity.

[0052] Optionally, the method is performed at least once when the three-phase motor is not energized and at least once when the three-phase motor is energized. This provides the advantage that both magnetization parameters k1 and k2 can be reliably determined.

[0053] Optionally, the method is iteratively performed alternately with the three-phase motor de-energized and energized, wherein at least a portion of the magnetization parameters determined during one execution of the method are used as predetermined estimates of the magnetization parameters during subsequent iterative executions of the method. This offers the advantage that the interdependence of magnetization parameters k1 and k2 can also be taken into account, thereby enabling a gradual improvement in the accuracy of the determination of both magnetization parameters. For example, the specification of magnetization parameter k1 performed during an iterative execution of the method can be taken into account in the further specification of magnetization parameter k2, and vice versa. In this way, a reliable determination of both magnetization parameters can be achieved.

[0054] Optionally, the rotation angles are predetermined as predetermined rotation angles at which the respectively measured voltage has a local maximum and / or local minimum. This offers the advantage that the magnetization parameters can be determined particularly reliably based on the measured values ​​of the measured voltage at the maximum and / or minimum values. This also offers the advantage that the magnetization parameters can be determined using a small number of measured values ​​or rotation angles based on these predetermined rotation angles. Optionally, at least a portion of the electrical revolutions of the rotor relative to the stator includes at least 90° and at least a predetermined rotation angle at which the measured voltage has a local minimum, and at least one further predetermined rotation angle at which the measured voltage has a local minimum and a local maximum. This offers the possibility of reducing the rotation angles, thereby shortening the measurement effort and duration.

[0055] Optionally, the predetermined rotation angle is predetermined so that the continuous course of the voltage over the electrical revolution can be determined based on the voltage measured at the predetermined rotation angle. This makes it easy to identify those rotation angles at which the measured voltage has a local minimum or local maximum. This can also be advantageous because it makes it easier to adapt the analog value to the measured value.

[0056] Optionally, a predetermined estimated value of the magnetization parameter is incorporated into the calculation of the simulation value via a predetermined mathematical model for the respective inductances of the first and second phases of the three-phase motor. In particular, the predetermined mathematical model can provide the possibility of automatically calculating the simulation value, for example by storing the predetermined mathematical model on the computing unit.

[0057] Optionally, the method further comprises pre-positioning the rotor relative to the stator to a predetermined starting rotation angle before the current pulse is fed. This can be particularly advantageous because the range of rotation angles over which the rotor rotates during the measurement covers the desired range and / or begins at the desired range. For example, the rotation angle range can be optimized in such a way as to maximize the probability of local maxima and / or minima occurring within the predetermined rotation angle range. This can also provide advantages for enabling and / or increasing the comparability of several different measurements. Optionally, the pre-positioning is performed by mechanically forced positioning and / or by applying current to the three-phase motor until the starting rotation angle is reached. This provides the opportunity for automated pre-positioning.

[0058] Furthermore, it is an object of the present invention to provide a method for reliably determining the initial rotor position of a three-phase electric motor with low hardware requirements, which in particular also allows a reliable north-south detection.

[0059] In another aspect, the present invention relates to a method for determining an initial rotor position of a three-phase motor using a control circuit. The method comprises energizing a first phase and a second phase of the three-phase motor with a current that increases or decreases during a first energization interval, and determining an induced voltage in an unenergized third phase of the three-phase motor at at least two measurement time points spaced apart in time. The method further comprises determining an induced voltage difference of a voltage induced in the third phase between two time-spaced induced measurement time points by taking into account a voltage change in a control circuit of the first phase and / or the second phase caused by energizing the first phase and the second phase during the first energization interval. The method further comprises determining an initial rotor position based on the determined induced voltage difference.

[0060] In another aspect, the invention relates to a control unit which is configured to determine an initial rotor position of a three-phase electric motor using the method according to the invention.

[0061] In another aspect, the invention relates to an electric motor comprising a stator and a rotor rotatable relative to the stator, wherein the electric motor is configured to determine an initial rotor position of the electric motor by means of a method according to the invention.

[0062] An electric motor comprises a stator and a rotor. In particular, the electric motor can be designed as a three-phase motor, having three phases or three windings, each of which can be energized via its own connection. The windings can be arranged in the motor so that they at least partially overlap. For example, the phases of the electric motor can be connected to one another in a delta connection or in a star connection. The terms "motor" and "motor" are used synonymously in this disclosure.

[0063] The rotor position corresponds to the rotation angle of the rotor relative to the stator. Knowledge of the rotor position may be particularly necessary for reliable commutation of the motor and for efficient operation of the motor. The rotor's rotation angle can range from 0° to 360°. Alternatively, the rotation angle can be assigned within a range from 0° to 180° in conjunction with determining the north-south orientation of the rotor or rotor magnets.

[0064] The inductive voltage divider can correspond in particular to the voltage drop between the connection of the de-energized phase and ground. This voltage is determined by the ratio of the inductances L1 and L2 of the first and second energized phases, which in turn depends on the position of the rotor.

[0065] When determining the induced voltage difference, taking into account the voltage change in the control circuit of the first phase and / or the second phase caused by the energization of the first phase and the second phase within the first energization interval means that the influence of the voltage change in the control circuit is determined or estimated and at least partially compensated for the induced voltage difference to be determined.

[0066] The present invention offers the advantage that possible influences and / or measurement errors in determining the induced voltage difference and influences and / or measurement errors in determining the initial rotor position due to voltage variations in the control circuit can be reduced or avoided. Consequently, this offers the advantage that the reliability of determining the initial rotor position can be increased because undesirable measurement errors can be reduced or avoided.

[0067] Furthermore, the present invention offers the advantage that, taking into account the voltage variations induced in the control circuit, absolutely no additional hardware is required to determine the initial rotor position. Specifically, the present invention offers the advantage that no separate sensor has to be provided to determine the initial rotor position, thereby allowing the manufacturing costs of the three-phase motor and the control circuit to be kept low.

[0068] Furthermore, the present invention offers the advantage that the initial rotor position can be reliably determined even for electric motors with low inductance, in particular for high-current three-phase electric motors, where, according to conventional methods, influences that typically occur lead to significant measurement errors and, accordingly, to a significantly reduced reliability of the determination of the initial rotor position. Consequently, the present invention offers the advantage that the method according to the invention for determining the initial rotor position is universally applicable to different types of three-phase electric motors, so that the number of control units and / or methods required for determining the initial rotor position of different types of three-phase electric motors can be kept low.

[0069] Optionally, the first and second phases are energized in such a way that the current increases or decreases strictly monotonically and optionally linearly during the first energization interval. This allows for reliable determination of the sign of the current change, i.e., whether the current increases or decreases during the energization interval. The sign of the current change determined in this manner can then be used for north-south detection of the rotor position.

[0070] Optionally, during the energization interval in the control circuit of the first and / or second phase, the voltage change caused by energizing the first and / or second phase corresponds to a shunt voltage difference, which is generated by the voltage drop across a shunt resistor used to measure the phase current in the first and / or second phase between two spaced-apart shunt measurement time points. The shunt voltage difference can optionally represent a primary cause of undesirable variations in the induced voltage difference, which can distort the determination of the induced voltage difference and, accordingly, reduce the reliability of the initial rotor position determination. The shunt voltage difference can optionally be very small, for example, in the range of a few millivolts. In many cases, if the induced voltage difference is very large compared to the shunt voltage difference, the effect of the shunt voltage difference on the initial rotor position determination is negligible. However, if the induced voltage difference is very small or even of the same order of magnitude as the shunt voltage difference, the shunt voltage difference can have a significant and undesirable effect on the determined induced voltage difference, thereby affecting the determination of the initial rotor position. However, by taking the shunt voltage difference into account when determining the induced voltage difference, these undesirable effects can be reduced or avoided.

[0071] Optionally, the inductive measurement time points include a first inductive measurement time point and a second inductive measurement time point, and the shunt measurement time points include a first shunt measurement time point and a second shunt measurement time point, wherein the first inductive measurement time point is no more than 5 μs away from the first shunt measurement time point, and / or the second inductive measurement time point is no more than 5 μs away from the second shunt measurement time point. This offers the advantage that the corresponding shunt voltage measurement and the inductive voltage measurement are performed as close as possible to one another, and accordingly, the shunt voltage difference and the inductive voltage difference are determined within approximately the same time interval. This allows for consideration of a shunt voltage difference that is as close as possible to the shunt voltage difference or the dominant shunt voltage that actually prevailed during the measurement of the inductive voltage difference or the plurality of inductive voltages. This enables particularly accurate compensation for the influence of the shunt voltage difference when determining the inductive voltage difference.

[0072] Optionally, the first inductive measurement time point and the second inductive measurement time point are separated from each other by at least 10 μs, optionally at least 100 μs, or optionally at least 1 ms. Alternatively or additionally, the first shunt measurement time point and the second shunt measurement time point are separated from each other by at least 10 μs, optionally at least 100 μs, or optionally at least 1 ms. This provides the advantage that the time period between the measurement time points of the induced voltage and / or shunt voltage is long enough to achieve a significant increase or decrease in the current flowing through the inductor and / or shunt, and thus to obtain a suitable signal amplitude for determining the induced voltage difference.

[0073] Optionally, the voltage change in the first phase and / or the second phase caused by the energization of the first phase and / or the second phase during the energization interval takes into account a predetermined, estimated shunt voltage difference. In other words, according to this alternative embodiment, the determination of the shunt voltage difference can be limited to estimating it rather than measuring it. Therefore, this alternative embodiment provides the possibility of considering the shunt voltage or shunt voltage difference by estimation rather than measurement or determination. For example, the estimation can be based on a calculation of an estimated value based on other known parameters and / or based on experimentally determined and provided data and / or empirical values. For example, the shunt voltage difference can be estimated using the following mathematical relationship:

[0074]

[0075] Here, ΔU 分流 represents the shunt voltage difference, U represents the supply voltage, L represents the inductance of the first or second phase (assuming the same), ΔT represents the time interval between the first and second induced voltage measurements, and R 分流 Denotes the ohmic resistance of the shunt resistor. Such an estimation of the shunt voltage difference may thus provide a solution, wherein the shunt voltage difference does not need to be determined or measured and the hardware and / or computational effort may thus be kept low.

[0076] Optionally, the voltage change in the first and / or second phases caused by energizing the first and / or second phases during the energization interval can be taken into account by reducing the induced voltage difference by a value proportional to the shunt voltage difference. For example, the value proportional to the shunt voltage difference can represent the value of the shunt voltage difference itself. However, according to other embodiments, fractions and / or multiples of the shunt voltage difference can also be used for this purpose. The fact that the induced voltage difference is reduced by a value proportional to the shunt voltage difference means that the difference between the induced voltage difference and the shunt voltage difference is used as the determined induced voltage difference for determining the initial rotor position. In this way, the influence of the shunt voltage difference on the determined induced voltage difference can be reduced or even completely eliminated.

[0077] Optionally, the voltage change in the first and / or second phases caused by the energization of the first and / or second phases during the energization interval is accounted for by reducing the induced voltage difference by half the value of the shunt voltage difference. This provides a particularly simple and reliable form of accounting. The change in the shunt voltage essentially affects the measured change in the induced voltage at the inductive voltage divider by dividing the error by the inverse voltage divider ratio of the inductive voltage divider. Since the voltage divider ratio at the inductive voltage divider is position-dependent, for accurate compensation, the voltage divider ratio can be actually measured and the shunt voltage difference divided by the measured voltage divider ratio can be subtracted from the induced voltage difference. Since in many embodiments, the voltage divider ratio at the inductive voltage divider is measured during rotor position detection, this can be achieved without an additional measurement step. However, to simplify and minimize the required computational power, a fixed voltage divider ratio of 1 / 2 can be used, so that half the shunt voltage difference is subtracted from the induced voltage difference. For many electric motors in use, the position dependence of the voltage divider ratio of the inductive voltage divider is only in the range of 1% to 10%. Therefore, a voltage division ratio of 1 / 2 can represent a very useful approximation that can be used as a basis for effectively considering the shunt voltage difference.

[0078] Optionally, the method for determining the initial rotor position further includes energizing the first and second phases of the three-phase motor with a current that increases or decreases during a second energization interval, wherein the voltage applied for energization has a voltage direction opposite to the voltage applied during the first energization interval; and determining an induced voltage in an unenergized third phase of the three-phase motor at at least two measurement time points spaced apart from each other during the second energization interval. Furthermore, the method according to this optional embodiment includes determining an induced voltage difference of the voltage induced in the third phase between two temporally spaced induced measurement time points during the second energization interval by taking into account voltage changes in the control circuit of the first and / or second phases caused by energization of the first and second phases during the second energization interval; and determining the initial rotor position based on the induced voltage difference determined during the first and second energization intervals. This provides the advantage of eliminating any measurement errors caused by any possible interference effects, as long as these interference effects occur equally in both energization intervals. This further improves the reliability of the initial rotor position determination. By using two opposing measurements, any inaccuracies caused by the hysteresis properties of the stator material can be reduced.

[0079] Optionally, before the first and / or second energization intervals, a voltage opposite to the voltage applied for energization is applied to the first and second phases for a predetermined time period. This offers the advantage that any influences caused by any remaining magnetization can be reduced or eliminated by the preceding measurement and / or energization. The predetermined time period may optionally not exceed 100 ms.

[0080] The features and embodiments mentioned above and explained below should be considered as disclosed not only in the combination explicitly mentioned in each case, but also in other technically reasonable combinations and in the disclosure of the embodiments. In particular, the various aspects of the present disclosure can be combined with one another. Thus, optionally, the method for providing a commutation interval may include the use of magnetization parameters determined using the method according to another aspect of the present disclosure. Alternatively or additionally, the method for determining the initial rotor position of a three-phase motor according to one aspect of the present disclosure may include the use of magnetization parameters determined using the method according to another aspect of the present disclosure. The method for providing a commutation interval according to one aspect of the present disclosure and the method for determining the initial rotor position of a three-phase motor according to one aspect of the present disclosure can also be implemented in a control device for a three-phase motor, which optionally uses the magnetization parameters of the three-phase motor determined according to the method according to another aspect of the present disclosure. In other words, the various aspects of the present invention can be combined with one another or used independently of one another.

[0081] Further details and advantages of the present invention will now be explained in more detail using the following embodiments and preferred embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 A schematic diagram of a three-phase electric motor with a control unit according to an alternative embodiment is shown.

[0083] Figure 2A and 2B Shows a star connection according to an alternative embodiment ( Figure 2A ) and triangle connection ( Figure 2B ) is an exemplary schematic diagram of a three-phase motor.

[0084] Figure 3 An exemplary initial commutation interval is shown.

[0085] Figure 4 Several processes of the measured voltage difference are shown.

[0086] Figure 5 An example of a measurement process of the voltage U3 across the inductive voltage divider is shown in the graph 100 .

[0087] Figure 6 An example of a comparison of the measured course of the voltage U3 at the inductive voltage divider with a simulated course of the voltage optimized for the magnetization parameter k1 is shown.

[0088] Figure 7 In the diagram 300 , the rotation angle is shown as The voltage across the inductive voltage divider is measured as a function of the .

[0089] Figure 8A comparison of simulated and measured values ​​for determining the voltage of k2 is shown.

[0090] Figure 9 Graphs 500 and 510 are shown here, which illustrate the linear dependence of the offset of the commutation interval on the operating current (i).

[0091] Figure 10 Shows the upper and lower switching limits g o,i and g u,i The correlation between the distances.

[0092] Figure 11 The curves show Figure 9 Correlation of commutation interval offset in .

[0093] Figure 12 Shown Figure 10 Different power supply voltage U S The parabolic course of the distance between the switching limits.

[0094] Figure 13 and Figure 14 The voltages are normalized to the supply voltage and correspond to Figure 11 and Figure 12 The curve of the curve.

[0095] Figure 15A and 15B Shows a star connection ( Figure 15A ) and triangle connection ( Figure 15B ) is a schematic diagram of a three-phase motor.

[0096] Figure 16 An exemplary time course of the voltages U1 , U2 and U3 and the current course through the second phase 2212 . 2 are shown;

[0097] Figures 17 to 20 Measured and simulated voltage U3 are shown as a function of the angle of rotation of the rotor relative to the stator.

[0098] Figure 21A and 21B Shows a star connection ( Figure 21A ) and triangle connection ( Figure 21B ) is a schematic diagram of a three-phase motor 10.

[0099] Figure 21C Shown is a schematic diagram of an electric motor and a control unit according to an alternative embodiment.

[0100] Figure 22A and 22B Shown is a conventional measurement and control circuit for a three-phase motor.

[0101] Figure 23 An exemplary process of determining the induced voltage is shown.

[0102] Figure 24 An exemplary process of inducing a voltage difference between the rotor and the stator throughout an electrical revolution is shown.

[0103] Figure 25 An exemplary course of the induced voltage in the third non-energized phase changing over time when the first and second phases are energized is shown.

[0104] Figure 26 The time course of the shunt voltage is shown.

[0105] Figure 27 The measurement signal of the uncorrected induced voltage difference is shown compared with half the value of the shunt voltage difference over the rotor rotation angle.

[0106] Figure 28 Examples of difference signals used for north-south detection or for determining initial rotor position are shown.

[0107] In the following drawings, for simplicity, the same or similar elements in various embodiments are denoted by the same reference numerals. DETAILED DESCRIPTION

[0108] Figure 1 A schematic diagram of a three-phase motor 10 according to an alternative embodiment is shown with a control unit 20. The control unit 20 is here communicatively connected to the three-phase motor 10 and is arranged to control the three-phase motor 10 and supply it with power.

[0109] Figure 2A and 2B Shows a star connection according to an alternative embodiment ( Figure 2A ) and triangle connection ( Figure 2B ) is an exemplary schematic diagram of a three-phase motor 10. The motor 10 has three connections 12.1, 12.2 and 12.3 for three different phases 12. Each of the three phases 12.1, 12.2 and 12.3 is characterized, for example, by an associated inductance L1, L2 or L3 and an associated ohmic resistance R1, R2 or R3. The supply voltage of the motor 10 is designated as U S Different voltages U1 , U2 and U3 can be present at the three connections 12 . 1 , 12 . 2 and 12 . 3 of the three phases, which also represent potential differences compared to ground potential.

[0110] Each of the three phases is connected at one end to the associated connection 12.1, 12.2 and 12.3. In the case of a star connection ( Figure 2A), the other end is connected to the star point 14 of the star connection. For delta connection ( Figure 2B ), and the other end is connected to the connector 12.1, 12.2 or 12.3 of the next phase respectively.

[0111] Figure 3 Exemplary initial commutation intervals k1, k2, and k3 are shown, which are selected for the rotation angle of the rotor relative to the stator of the three-phase motor 10. (in degrees) to generate high torques (M1, M2, M3) (arbitrary units). The switching intervals k1 to k3 are defined by the upper switching limit g o,0 and switching lower limit g u,0 Defined. The voltage or voltage difference between two consecutive measured values ​​of the measuring pulses with different polarity measured back in the respective commutation interval k1 to k3 is compared with the switching limit of the respective initial commutation interval k1 to k3. If the measured voltage U1, U2 or U3 reaches the switching limit g o,0 or g u,0 The commutation is changed so that the phases of the motor are switched. Two phases of the rotor of the motor are energized to operate at the highest torque M1, M2 or M3.

[0112] Figure 4 The respective non-energized phase 12.1, 12.2 or 12.3 is shown at a rotation angle The graphs above show several curves for the measured voltage differences U1, U2, and U3 based on the operating current i energizing the other two phases. The curves for the voltage differences U1, U2, and U3 of the corresponding de-energized phases are plotted for different operating currents i, with the operating current i varying between 0 and 1500 (arbitrary units). The graphs show that the intensity of the operating current i influences the course of the corresponding voltage differences U1, U2, and U3. Due to the asymmetry observed compared to low operating currents and the absence of current, high operating currents lead to changes in the amplitude and angular course of the voltage differences U1, U2, and U3. In the example shown, the amplitude of the voltage differences increases with increasing operating current, and the asymmetry increases. After comparing these voltage difference values ​​with the switching limits of the commutation intervals k1, k2, and k3 and performing commutation based on this comparison, these deviations have a direct impact on the commutation and, therefore, the function and efficiency of the motor.

[0113] Such deviations can be reduced or avoided by the method according to the invention for providing the commutation intervals. An exemplary method is explained below, which includes adapting the commutation intervals taking into account the operating current. However, the invention is not limited to the explained example.

[0114] Knowledge of the magnetizing parameters k1 and k2 and the maximum inductance of the individual phases is advantageous for adapting the commutation intervals.If the magnetizing parameters k1 and k2 are not known in advance, they can be determined, for example, using the following method.

[0115] A method for determining the magnetization parameters of electric motor 10 according to an alternative embodiment will be explained below with reference to the accompanying drawings. For illustrative purposes, phases 12.1 and 12.2 of electric motor 10 represent the first and second phases of electric motor 10 into which current pulses are fed according to this method, while the third, non-energized phase is formed by phase 12.3. However, the distribution or sequence can also be selected differently, depending on the requirements.

[0116] First, according to the method, the magnetization parameters k1 and k2 of the three-phase motor 10 are determined if these parameters are not already known. In addition, the method may include determining the maximum inductance L of the respective phases 12.1, 12.2 and 12.3. max (If these are not already known.) According to some optional embodiments, it can be assumed that the maximum inductance L of each phase max The same, so only the maximum inductance L of a single phase needs to be determined max In particular, L max It can be determined by measuring the inductance of one or both phases simultaneously in the appropriate ratio.

[0117] To determine the magnetization parameters of a three-phase motor, the voltage or voltage difference U3 at the connection representing the third phase 12.3 of the inductive voltage divider is measured during a measurement period. Multiple measurements are performed periodically or irregularly during the measurement period. During the measurement period, the rotor also rotates uniformly relative to the stator for at least a fraction of an electrical revolution. During the measurement period, the voltage is measured at the inductive voltage divider at least when the rotor assumes one of a plurality of predetermined rotational angles relative to the stator.

[0118] The rotation is performed at least over a portion of a complete electrical revolution so that at least a maximum and at least a minimum value of the measured voltage U3 are measured at the inductive voltage divider within the covered rotation angle range. Alternatively, the measurements can be performed at much smaller time intervals or rotation angle intervals, so that the course of the voltage U3 can optionally be reconstructed as a function of the rotation angle, although this is not absolutely necessary.

[0119] To determine magnetization parameter k1, voltage U3 is measured across an inductive voltage divider as a function of the rotational angle while the motor is not energized, i.e., without the voltage pulses used to generate the current pulses for determining the magnetization parameter. To determine magnetization parameter k2, a corresponding measurement is performed while motor 10 is energized, wherein, in addition to the current pulses, commutation of motor 10 is performed, wherein the commutation remains unchanged during the measurement and no adaptation to the changed rotational angle is performed. Both measurements can be performed multiple times, in particular alternatingly, in order to determine both magnetization parameters within the context of an iterative process.

[0120] The following first explains how to determine the magnetization parameter k1. To this end, motor 10 is controlled to periodically and repeatedly feed current pulses to first phase 12.1 and second phase 12.2 during a measurement period, while the rotor rotates slowly and evenly relative to the stator. Rotation can be performed manually by the user or automatically. Furthermore, while feeding the current pulses, the voltage value of voltage U3 is measured across an inductive voltage divider and the measured value is stored. It may be sufficient to store only the voltage values ​​of voltage U3 at those rotational angles where voltage U3 has a local minimum or maximum, and then assign these values ​​to the corresponding angle values. Alternatively, more frequent recording of the voltage values ​​can be used to identify and / or reconstruct the course of voltage U3 over the rotational angle range.

[0121] Figure 5 The graph 100 shows the rotation angle at the inductive voltage divider (ie at the connection of the third phase 12.3) when the motor is not energized. An example of a measurement process for voltage U3 during a process, i.e., when the motor is not energized in addition to the voltage pulses for generating the current pulses used to determine the magnetization parameter k1. The horizontal axis plots the rotation angle of the rotor relative to the stator from 0° to 360°, and the vertical axis plots the measured voltage difference in volts between the measured voltage values ​​of the measurement pulses of different polarity. Graph 100 shows that measured voltage U3 or the voltage difference has a sinusoidal curve with a period of approximately 180° and an amplitude of approximately 1.3 V, oscillating around the zero line.

[0122] Furthermore, the method comprises calculating a corresponding analog value of the voltage at the voltage divider at a predetermined rotation angle using a predetermined estimated value of the magnetization parameter and adapting the predetermined estimated value of the magnetization parameter so that the deviation of the analog value of the voltage from the measured value of the measured voltage U3 is minimized. For example, assuming that the values ​​k1=0 and k2=0 can be used as initial estimates for k1 and k2.

[0123] The analog values ​​are calculated using a given mathematical model. An exemplary mathematical model is explained below, but the present invention is not limited thereto. The mathematical model is described for the aforementioned example case, in which current pulses are supplied to first phase 12.1 and second phase 12.2, while voltage U3 or a voltage difference is measured on the unenergized third phase 12.3. This mathematical model is based on the following system of differential equations.

[0124] (3)

[0125] (4)

[0126] in

[0127] (5)

[0128] (6)

[0129] (7)i L1 =-i L2

[0130] Here, L1 and L2 represent the inductance of the first phase 12.1 and the second phase 12.2 of the three-phase motor, respectively. S denotes the supply voltage, R1 and R2 denote the ohmic resistance of the first phase 12.1 and the second phase 12.2, respectively, t denotes time and U3 denotes the voltage or voltage difference measured at the connection of the third phase 12.3. Assuming a maximum inductance L max is the same for all phases of the motor and can be determined, for example, by measuring the inductance of a single phase. If this is not possible due to the spatial overlap of the two phases, the maximum inductance of the two phases 12.1 and 12.2 can also be measured together and the value of the single inductance L can be determined by appropriate reduction max .Rotation angle Here, it is the angle of rotation of the rotor relative to the stator. The parameter k1 corresponds to the current-independent magnetization parameter, which characterizes the inductance change caused by the rotor magnets, and k2 corresponds to the current-dependent magnetization parameter, which characterizes the inductance change caused by the current supplied to the corresponding phase.

[0131] Using equations (3) to (7) presented above, it is then possible to calculate a simulated value for voltage U3 that corresponds to the expected measured voltage value at a predetermined rotation angle of the rotor during the measurement. Thus, voltage values ​​are calculated specifically for those rotation angles at which voltage U3 is also measured. These calculated voltage values ​​can then be compared with the measured voltage values. Based on this, the magnetization parameter k1 can be varied and optimized in a regression process, and the effect on the calculated simulated values ​​can be examined. For the initial simulation and optimization of k1 without energizing the motor, a predetermined initial value for k2, for example k2=0, can first be assumed. If k2 has already been determined and / or optimized in a previous measurement, this value can optionally be used.

[0132] In this way, an optimization of the magnetization parameter k1 can be performed, the aim of which is to minimize the deviation of the calculated voltage simulation value from the corresponding measured voltage value. The value of the magnetization parameter k1 determined in this way by simulation (with the smallest deviation) can then be assigned to the measured three-phase motor as the actual magnetization parameter k1 and / or used to determine the parameter k2.

[0133] Figure 6 An example is shown Figure 3 The measurement process of the voltage U3 on the inductive voltage divider (graph 100) versus the rotation angle Comparison of the simulated course of voltage U3 on φ1, optimized for magnetization parameter k1 (graph 200). It can be seen here that graph 100 and graph 200 have only slight deviations, so that a very precise adaptation of the simulation to the measured values ​​can be achieved by varying k1.

[0134] The corresponding method steps can then be carried out to determine the magnetization parameter k2 , wherein, in contrast to the method described above for determining the magnetization parameter k1 , the method steps are carried out while the motor is energized, with a defined, predetermined current and with a predetermined, constant commutation.

[0135] Figure 7 In the diagram 300 , the rotation angle is shown as The measurement process of the voltage U3 (vertical axis, in volts) across the inductive voltage divider as a function of the current (in degrees) is shown in graph 300 . Graph 300 corresponds to the course of voltage U3, which is generated by periodically repeating current pulses fed into first phase 12 . 1 and second phase 12 . 2 while the rotor rotates slowly and evenly relative to the stator during the measurement period. The measured voltage value U3 is then stored in a memory, which can be part of the motor control or designed separately from the motor. Graph 300 also shows that voltage U3 has local minima and maxima over one electrical revolution, which can be used to determine magnetization parameter k2.

[0136] When determining the magnetization parameter k2, the corresponding simulation value is also calculated and optimized and compared with the measured value, such as Figure 8 As shown, wherein the optimization is performed by changing the magnetization parameter k2. Figure 7 The graph of the measured voltages shown in FIG400 corresponds to the simulation and optimization process of the voltage U3. When optimizing the magnetization parameter k2, a predetermined value for k1 can be assumed. If the magnetization parameter k1 has already been determined and / or optimized in a previous measurement, this value can also be optionally used when determining the magnetization parameter k2. The accuracy of the determination of the magnetization parameters k1 and k2 can be improved by multiple iterative optimizations of the magnetization parameters k1 and k2, wherein each iterative step uses the result of the previous iterative step. Figure 8 As can be seen in FIG. 1 , a very good agreement between the simulated voltage value and the measured voltage value U3 can also be achieved for determining the magnetization parameter k2 .

[0137] Using the known maximum inductance L max With the known inductance parameters k1 and k2, the further determination of the adapted commutation interval can then be performed. Here, the operating current (i) occurring during operation of the three-phase motor is determined. The operating current (i) can be determined, for example, by suitable current measurements and / or by simulations.

[0138] For example, adapting the commutation interval according to the operating current can aim to redefine the upper and lower switching limits g o,i and g u,i , then determine the switching limit g with the initial commutation interval o,0 and g u,0 For example, the switching limits can be set so that the switching interval covers the rotation angle range 60° and the zero crossing of the voltage difference U1, U2 or U3 is located at the commutation interval 60° or the rotation angle range covered The resulting switching limit g is determined based on the determined operating current o,i and g u,i , which can then be used for commutation at the corresponding operating current i.

[0139] In addition, according to the switching limit g o,i and g u,i The voltage or voltage difference U1, U2 or U3 at the inductive voltage divider is calculated using equations (3) to (7) in the above differential equation system based on the operating current (i). This can be done digitally, for example. Example results of the voltage or voltage difference U1, U2 or U3 calculated based on the operating current (i) are shown in Figure 1. Figure 9 and Figure 10 shown. Figure 9Graphs 500 and 510 are shown, which illustrate the linear dependence of the offset of the commutation interval on the operating current (i) as a result of a simulation (graph 500) and after optimization using regression (graph 510). Graphs 500 and 510 show very high agreement and almost overlap. The offset has a linear course and can be described as a straight line using a straight line equation.

[0140] (8)U=m·i+n

[0141] In the example shown, the slope of this line is approximately m=0,15V / A, and the zero offset is n=0V.

[0142] The upper and lower switching limits corresponding to the voltage difference g o，i and g u，i The distance relationship between Figure 10 As shown, graph 600 corresponds to the simulation results and graph 610 corresponds to the results after optimization using the regression method. This quadratic correlation can be expressed as a parabola according to the following equation:

[0143] (9)AU=a·(i-i0) 2 +b

[0144] Here, ΔU represents the switching upper and lower limits g o，i and g u，i The voltage difference between the two, the parameter a represents the opening of the parabola, i0 represents the operating current of the initial commutation interval, which is not necessarily zero, especially in the case of average value measurement pulses, and the parameter b represents the peak offset. These values ​​a = 0,0062 V / A 2 , i0=0, and b=2,4V from graph 610.

[0145] Specifically, the parameters m, n, a, b, and i0 can be optimized as part of the optimization process to, for example, minimize the deviation of the calculated results of equations (8) and (9) from the simulated results of equations (8) and (9). Alternatively, the parameters m, n, a, b, and i0 can be optimized as part of the optimization process to minimize the deviation of the calculated results of equations (8) and (9) from the measured values.

[0146] Based on the determined and possibly optimized parameters m, n, a, b, and i0, the initial commutation intervals can then be adapted according to the operating current i. For example, the parameters m, n, a, b, and i0 can be transmitted to a control unit for controlling the electric motor and used by it to determine a suitable adaptation and offset of the commutation intervals based on the operating current i. Alternatively or additionally, for example, adapted commutation intervals for different operating currents can be predefined and stored in the control unit, for example, in the form of a table, so that the control unit can select and apply the corresponding assigned adapted commutation intervals according to the operating current i.

[0147] According to some embodiments, the power supply voltage or operating voltage U used for the operation of the motor is s Variations and / or fluctuations in may also have an impact on the commutation and, therefore, on the operation and efficiency of the motor. According to some embodiments, the supply voltage U may also be taken into account when adapting the commutation intervals. s fluctuations and / or changes in the

[0148] Figure 11 Multiple curves show the Figure 9 The dependence of the commutation interval offset on the curve with the minimum slope corresponding to a supply voltage U of 6 V s The curve with the largest slope corresponds to a power supply voltage U of 80V. s Accordingly, Figure 12 Shown in Figure 10 For different power supply voltages U s The parabolic course of the distance between the commutation limits, where the lower curve corresponds to a supply voltage U of 6V S , the top curve corresponds to the power supply voltage U of 80V S This shows that the power supply voltage U S Deviations from the expected supply voltage value (which forms the basis for the initial or adapted commutation intervals based on the operating current i) can also have a significant impact on the operation and efficiency of the motor, and it may be advantageous to take the supply voltage into account when adapting the commutation intervals.

[0149] In particular, the supply voltage U can be considered in the following way S , that is, the measured voltage or voltage difference U1, U2 or U3 is normalized to the power supply voltage, and the normalized voltage ΔU rel is used to select the commutation interval with appropriately normalized commutation interval switching limits. Therefore, the voltage used can be determined as follows:

[0150]

[0151] Power supply voltage U SThis can be determined, for example, by one or more corresponding voltage measurements. For example, the supply voltage can be measured at start-up of the motor and / or periodically or irregularly during operation. This also offers the advantage of simplifying the use of the motor and its control in networks with different voltages. For example, the motor can then be used in a 12V, 24V, or 48V vehicle electrical system, and the commutation intervals can be adapted accordingly.

[0152] Figure 13 and Figure 14 The voltages are normalized to the supply voltage and correspond to Figure 11 and Figure 12 The results show that the normalization can significantly reduce the dependence of the offset and distance between commutation interval boundaries on the supply voltage. This can further improve the reliability and efficiency of the motor and its control.

[0153] Figure 15A and 15B Shows a star connection ( Figure 15A ) and triangle connection ( Figure 15B ) is an exemplary schematic diagram of a three-phase motor 210. The motor 210 has three connections 2212.1, 2212.2 and 2212.3 for three different phases 212. Each of the three phases 212.1, 212.2 and 212.3 is characterized, for example, by an associated inductance L1, L2 or L3 and an associated ohmic resistance R1, R2 or R3. The supply voltage of the motor 210 is denoted by U S Different voltages U1 , U2 and U3 can be present at the three connections 212 . 1 , 212 . 2 and 212 . 3 of the three phases, which also represent potential differences compared to ground potential.

[0154] Each of the three phases is connected at one end to the associated connection 212.1, 212.2 and 212.3. In the case of a star connection ( Figure 15A ), the other end is connected to the star point 214 of the star connection. For the delta connection ( Figure 15B ), and the other end is connected to the connector 212.1, 212.2 or 212.3 of the next phase respectively.

[0155] A method for determining the magnetization parameters of motor 210 according to an alternative embodiment will be explained below with reference to the accompanying drawings, although the method is not limited to this alternative embodiment. For purposes of illustration, phases 212.1 and 212.2 of motor 210 represent the first and second phases of motor 210 into which current pulses are fed according to the method, while the third, non-energized phase is formed by phase 212.3. However, the distribution or sequence may be selected differently as desired.

[0156] Figure 16 An exemplary time course of the voltages U1, U2 and U3 as a function of time t and a current course i of the second phase 212.2 as a function of time t are shown. L2 As can be seen in the top two graphs representing U1 and U2, the first phase and the second phase are alternately provided with voltage pulses so that alternating rising and falling current pulses are fed to the second phase 212.2 (i L2 , Figure 16 bottom graph). Figure 16 Also shown in the third diagram is the time course of the voltage at the connection of the third phase or third phase 212.3, which is assumed to be higher or lower than the supply voltage U S 2, depending on whether the voltage is applied to the first phase 212.1 or the second phase 212.2. In addition, a voltage rise can be seen in each alternating interval of the voltage U3, the amplitude of which is small compared to the total voltage U3, which is caused by the current change during this period.

[0157] The current i through the second phase 212.2 L2 The voltage pulse is applied and the current i is maintained accordingly. L2 The slope of the time interval in Figure 16 The curve diagram is marked as t1.

[0158] When measuring voltage U3 on the third, non-energized phase, a voltage difference between the voltages at different times can also be evaluated, where the difference in voltage U3 is formed at the time when a voltage pulse is fed to the first phase, and voltage U3 is formed at a second time when a voltage pulse is fed to the second phase 212.2. Thus, voltage U3 or the aforementioned voltage difference can be used to perform the method. Within the scope of this disclosure, explanations regarding voltage U3 also include corresponding explanations using the voltage difference.

[0159] To determine the magnetization parameters of a three-phase motor, during a measurement, the voltage or voltage difference at the connection of the third phase 212.3, representing the contact terminal of the inductive voltage divider, is measured. Multiple measurements are performed periodically or irregularly during the measurement. During the measurement, the rotor also rotates uniformly relative to the stator by at least a fraction of an electrical revolution. During the measurement, the voltage is measured at the inductive voltage divider at least when the rotor assumes one of a plurality of predetermined rotational angles relative to the stator.

[0160] The rotation is performed at least over a portion of a complete electrical revolution so that at least a maximum and at least a minimum value of the measured voltage U3 are measured at the inductive voltage divider within the covered rotation angle range. Alternatively, the measurements can be performed at much smaller time intervals or rotation angle intervals, so that the course of the voltage U3 can optionally be reconstructed as a function of the rotation angle, although this is not absolutely necessary.

[0161] To determine magnetization parameter k1, voltage U3 is measured across an inductive voltage divider as a function of the rotational angle while the motor is not energized, i.e., without the voltage pulses used to generate the current pulses for determining the magnetization parameter. To determine magnetization parameter k2, a corresponding measurement is performed while motor 210 is energized, wherein, in addition to the current pulses, commutation of motor 210 is performed, the commutation remaining unchanged during the measurement and no adaptation to the changed rotational angle is performed. Both measurements can be performed multiple times, in particular alternatingly, to determine both magnetization parameters within the context of an iterative process.

[0162] The determination of magnetization parameter k1 will be explained below. For this purpose, current pulses are periodically and repeatedly fed to first phase 212.1 and second phase 212.2 via the control of motor 210 during the measurement period, while the rotor rotates slowly and evenly relative to the stator during the measurement period, or individually between consecutive measurements. The rotation can be performed manually by the user or automatically. Furthermore, while feeding the current pulses, the voltage value of voltage U3 is measured across an inductive voltage divider and the measured value is saved. It may be sufficient to store only the voltage values ​​of voltage U3 at those rotation angles where voltage U3 has a local minimum or maximum, and then assign these values ​​to the corresponding angle values. Alternatively, the voltage values ​​can be recorded more frequently, allowing the course of voltage U3 over the rotation angle range to be identified and / or reconstructed.

[0163] Figure 17 Graph 2100 shows the voltage at the inductive voltage divider (i.e., the connection of the third phase 212.3) as the rotation angle increases when the motor is not energized, i.e., when the motor is not energized except for the voltage pulses used to generate the current pulses to determine the magnetization parameter k1. Graph 2100 illustrates an example of a measurement process for a changing voltage U3. The horizontal axis plots the rotation angle of the rotor relative to the stator from 0° to 360°, and the vertical axis plots the measured voltage difference in volts between the measured voltage values ​​of measuring pulses of different polarity. Graph 2100 shows that measured voltage U3 or the voltage difference has a sinusoidal curve with a period of approximately 180° and an amplitude of approximately 1.3 V, oscillating around the zero line.

[0164] Furthermore, the method comprises calculating a corresponding analog value of the voltage at the voltage divider at a predetermined rotation angle using a predetermined estimated value of the magnetization parameter and adapting the predetermined estimated value of the magnetization parameter such that deviations of the analog value of the voltage from the measured value of the measured voltage U3 are minimized.

[0165] The analog values ​​are calculated using a given mathematical model. An exemplary mathematical model is explained below, but the present invention is not limited thereto. The mathematical model is described for the aforementioned example case, in which current pulses are supplied to first phase 212.1 and second phase 212.2, while voltage U3 or a voltage difference is measured on third, non-energized phase 212.3. This mathematical model is based on the following system of differential equations.

[0166] (10)

[0167] (11) in

[0168] (12)

[0169] (13)

[0170] (14)i L1 =-i L2

[0171] Here, L1 and L2 represent the inductance of the first phase 212.1 and the second phase 212.2 of the three-phase motor, respectively. S denotes the supply voltage, R1 and R2 denote the ohmic resistance of the first phase 212.1 and the second phase 212.2, respectively, t denotes time and U3 denotes the voltage or voltage difference measured at the connection of the third phase 212.3. Assuming the maximum inductance L max is the same for all phases of the motor and can be determined, for example, by measuring the inductance of a single phase. If this is not easy to achieve due to the spatial overlap of the two phases, the maximum inductance of the two phases 212.1 and 212.2 can also be measured together and the value of the single inductance L can be determined by appropriately scaling down max .Rotation angle Here, it is the angle of rotation of the rotor relative to the stator. The parameter k1 corresponds to the current-independent magnetization parameter, which characterizes the inductance change caused by the rotor magnets, and k2 corresponds to the current-dependent magnetization parameter, which characterizes the inductance change caused by the current supplied to the corresponding phase.

[0172] Using equations (10) to (14) presented above, it is then possible to calculate a simulated value for the voltage U3 that corresponds to the expected measured voltage value at a predetermined rotation angle of the rotor during the measurement. Thus, voltage values ​​are calculated specifically for those rotation angles at which the voltage U3 is also measured. These calculated voltage values ​​can then be compared with the measured voltage values. Based on this, the magnetization parameter k1 can be varied and optimized in a regression process, and the effect on the calculated simulated values ​​can be examined. For the initial simulation and optimization of k1 without powering the motor, a predetermined value for k2 can first be assumed. If k2 has already been determined and / or optimized in a previous measurement, this value can optionally be used. Alternatively, for example, it is possible to initially start and / or perform a simulation or optimization with k2=0.

[0173] In this way, an optimization of the magnetization parameter k1 can be performed, the aim of which is to minimize the deviation of the calculated voltage simulation value from the corresponding measured voltage value. The value of the magnetization parameter k1 determined in this way by simulation (with the smallest deviation) can then be assigned to the measured three-phase motor as the actual magnetization parameter k1 and / or used to determine the parameter k2.

[0174] Figure 18 An example is shown Figure 17 The measurement process of the voltage U3 on the inductive voltage divider in FIG. 2100 as a function of the rotation angle Comparison of the simulated course of voltage U3 on φ1, optimized for magnetization parameter k1 (graph 2200). It can be seen here that graph 2100 and graph 2200 have only slight deviations, so that a very precise adaptation of the simulation to the measured values ​​can be achieved by varying k1.

[0175] The corresponding method steps can then be carried out to determine the magnetization parameter k2 , wherein, in contrast to the method described above for determining the magnetization parameter k1 , the method steps are carried out while the motor is energized, with a defined, predetermined current and with a predetermined, constant commutation.

[0176] Figure 19 The graph 2300 shows the voltage U3 (vertical axis, in volts) across the inductive voltage divider as a function of the rotation angle The measurement process of the change in ΔV (in degrees) is shown. Graph 2300 corresponds to the course of voltage U3, which is generated by periodically repeating current pulses fed into first phase 212.1 and second phase 212.2, while the rotor rotates slowly and evenly relative to the stator during the measurement. The measured voltage value U3 is then stored in a memory, which can be part of the motor control or designed separately from the motor. Graph 2300 also shows that voltage U3 has local minima and maxima over one electrical revolution, which can be used to determine magnetization parameter k2.

[0177] When determining the magnetization parameter k2, the corresponding simulation value is also calculated and optimized and compared with the measured value, such as Figure 20 As shown, wherein the optimization is performed by changing the magnetization parameter k2. Figure 19 The graph of the measured voltages shown in FIG2400 corresponds to the simulation and optimization process of the voltage U3. When optimizing the magnetization parameter k2, a predetermined value for k1 can be assumed. If the magnetization parameter k1 has already been determined and / or optimized in a previous measurement, this value can also be optionally used when determining the magnetization parameter k2. The accuracy of the determination of the magnetization parameters k1 and k2 can be improved by multiple iterative optimizations of the magnetization parameters k1 and k2, wherein each iterative step uses the result of the previous iterative step. Figure 20 As can be seen in Figure 1, a very good agreement between the simulated voltage value U3 and the measured voltage value U3 can also be achieved for determining the magnetization parameter k2. The first initial value can be selected, for example, as k1=k2=0, and after each iteration step, the last determined optimized value can be used for the other parameters respectively.

[0178] Figure 21A and 21B Shows a star connection according to an alternative embodiment ( Figure 21A ) and triangle connection ( Figure 21B ) is an exemplary schematic diagram of a three-phase motor 310. The motor 310 has three connections 312.1, 312.2 and 312.3 for three different phases 312. Each of the three phases 312.1, 312.2 and 312.3 is connected, for example, via an associated inductance L1, L2 or L3 and an associated ohmic resistor R M1 、R M2 or R M3 The power supply voltage of the motor 310 is marked as U S Different voltages U1 , U2 and U3 can be present at the three connections 312 . 1 , 312 . 2 and 312 . 3 of the three phases, which also represent potential differences compared to ground potential.

[0179] Each of the three phases is connected at one end to the associated connection 312.1, 312.2 and 312.3. In the case of a star connection ( Figure 21A ), the other end is connected to the star point 314 of the star connection. For the delta connection ( Figure 21B ), and the other end is connected to the connector 312.1, 312.2 or 312.3 of the next phase respectively.

[0180] Figure 21C The figure shows an electric motor 310 having a control unit 320, which are communicatively connected to one another. The control unit 320 is designed separately from the electric motor 310 and is configured to provide control signals thereto and, in particular, to commutate the electric motor 310. Furthermore, the control unit is configured to determine an initial rotor position of the electric motor 310.

[0181] Figure 22A A conventional measurement and control circuit 3100 for a three-phase motor 310 is shown. Reference numerals M1 to M6 denote field effect transistors (FETs) for applying up to three cycle voltages to the three-phase motor. The arrangement of OPV IC1 and resistors R2 to R5 represents an amplifier network for amplifying the voltage drop across shunt resistor R1 and making it available to a control unit.

[0182] Resistors R6 to R 10 Here, it serves as a resistive voltage divider, which divides the phase voltages U1 , U2 and U3 to be measured for determining the initial position of the rotor angle on an inductive voltage divider and then supplies the divided voltage to a control unit.

[0183] Shunt resistor R1 is used to measure the individual phase currents, i.e. the currents supplied to or flowing into the individual phases of a three-phase motor. The shunt voltage dropped across shunt resistor R1 is typically small compared to the other voltages dropped in the control circuit and can be, for example, in the double-digit millivolt range.

[0184] Figure 22BAnother conventional measurement and control circuit 3102 for a three-phase motor 310 is shown. The circuit has multiple shunt resistors, namely, shunt resistors R1 and R2, each connected in series with a FET M1 or M3 for energizing one of the three-phase motor phases. Although the three-phase motor 310 has three phases, in this configuration, only two shunt resistors, R1 and R2, are typically sufficient to determine the current in the corresponding phase, since the currents of the other two phases can be calculated from the measured currents via a node group, resulting in the third phase current of the motor 310. Using such a control circuit 3102, the voltage drop across both shunt resistors R1 and R2 can optionally be considered when determining the induced voltage difference. The shunt voltage difference is optionally considered or compensated for only during those energization intervals where a shunt resistor R1 or R2 is actually present in the current path. However, if energization is performed such that the current is dissipated to ground via the external right path, i.e., via FET M5, no compensation is required in this case. Therefore, it may be appropriate to take the shunt voltage difference into account for some power-on intervals but not for other power-on intervals. Optionally, this may also be taken into account when averaging multiple current supply intervals to improve the signal.

[0185] The following describes a method for using an alternative embodiment of the present invention. Figure 22A The method by which the control circuit determines the initial rotor position of a three-phase motor is shown and its background is explained.

[0186] Figure 23 An exemplary course of a voltage induced in a non-energized third phase of a three-phase motor 310 is shown, the voltage being determined as a function of the rotor position or rotation angle relative to the stator in degrees (horizontal axis) over a complete electrical revolution, during which the first and second phases are energized. An induced voltage difference is plotted on the vertical axis, which is generated by determining the induced voltage in the third non-energized phase or the induced voltage across an inductive voltage divider at at least two temporally spaced measurement points in time during a first energization interval, during which the first and second phases are energized with a rising or falling current. Figure 23 The graph in FIG shows that the induced voltage difference depends largely on the rotor position, and in particular changes sign depending on the rotor position. Thus, by determining the induced voltage difference, the initial rotor position and the north-south orientation of the three-phase motor 310 can be determined. The rotor magnetic pole position can be determined based on the sign of the induced voltage difference, which is generated by determining the induced voltage difference at the main rotor position to be determined.

[0187] Figure 24An example of the course of the induced voltage difference during the entire electrical revolution of the rotor relative to the stator of a three-phase motor 310 having a low current dependence of inductance is shown. The low current dependence of inductance means that the course of the induced voltage difference no longer has a sign change, and therefore, it is more likely that the north-south orientation of the rotor and the initial rotor position cannot be reliably determined based on the sign of the induced voltage difference (without further correction). Figure 24 As can be seen in Figure 1, the values ​​of the induced voltage difference range from about 0.015 V to about 0.07 V and are only assumed to be positive. This is further reinforced by the fact that in some motors the magnitude of the voltage difference is even lower, further increasing the probability that no sign change occurs.

[0188] although Figure 24 The process of inducing a voltage difference in Figure 23 The process in the rotor is described, and the deviations consist primarily of vertical offsets. However, this means that the initial rotor position cannot be reliably determined based on this during operation. This is because, when determining the initial rotor position, it is not possible to average the induced voltage differences, as the induced voltage differences must be determined over the entire electrical revolution, or at least a large portion of it. However, this is not possible when the rotor is stationary and in order to determine the initial rotor position, as the rotor position naturally does not change when the rotor is stationary.

[0189] According to the explained embodiment, the cause of the offset is taken into account and is taken into account when determining the induced voltage difference to determine the initial rotor position.

[0190] When using according to Figure 22A The offset occurs when using a control circuit. To determine the initial rotor position, the first and second phases of three-phase motor 310 are energized with a current that increases or decreases during a first energization interval. This current change, in turn, causes a voltage change in the voltage drop across shunt resistor R1. This voltage drop affects the induced voltage difference determined between two interval induction measurement time points during the first energization interval, resulting in the observed offset. In particular, in the case of a three-phase motor where the current only slightly alters the magnetic field when the first and second phases are energized, thus resulting in only a small induced voltage and induced voltage difference, the voltage change in the shunt voltage resulting in the shunt voltage difference at two intervals of shunt measurement time points as close as possible to the induction measurement time point in the energization interval can have similar characteristics, magnitude, or amplitude to the induced voltage difference. In this case, the voltage change in the voltage drop across the shunt resistor, or the shunt voltage difference, can significantly offset the induced voltage difference and, in particular, cause the sign of the induced voltage difference to no longer change depending on the rotor position.

[0191] Figure 25An example of the time course of the induced voltage in the third, unenergized phase is shown, while the first and second phases are energized with increasing current. The supply voltage is 5V in this case. The time window of approximately 25 μs shown here can be considered an exemplary energization interval, during which the induced voltage in the third phase (i.e., at the inductive voltage divider) increases by approximately 100 mV. If two inductive measurement points are set at the beginning and end of the illustrated time window or exemplary energization interval, an induced voltage difference of approximately 100 mV results.

[0192] Figure 26 Shows that for Figure 25 The time course of the shunt voltage in the same time window, i.e. during the power-on period across the shunt resistor R1 ( Figure 22A ) across the shunt voltage. It can be seen that the shunt voltage increases by about 130mV during the energizing interval, so the shunt voltage difference between the beginning and end of the energizing interval itself appears to be greater than the induced voltage difference ( Figure 25 Since the shunt voltage or shunt voltage difference is divided by the error in the inverse voltage divider ratio of the sensing voltage divider, it affects the voltage change measured at the sensing voltage divider. If the shunt voltage difference is not taken into account, this can lead to errors in the measurement of the sensed voltage difference. Therefore, according to the explained embodiment, the shunt voltage difference is taken into account when determining the sensed voltage difference.

[0193] According to an alternative embodiment, when determining the induced voltage difference by measuring the voltage divider ratio of the inductive voltage divider (which in turn depends on the rotor position), the shunt voltage difference can be taken into account and then subtracted from the induced voltage difference by dividing the shunt voltage difference by the measured voltage divider ratio. This has the advantage that the effect of the shunt voltage difference on the induced voltage difference can be accurately accounted for and compensated for. Because the voltage divider ratio can be measured at the inductive voltage divider while determining the initial rotor position, this can be accomplished without additional effort in some alternative embodiments.

[0194] According to another alternative embodiment, the shunt voltage difference can be accounted for differently when determining the induced voltage difference. According to this alternative embodiment, a fixed voltage division ratio of 2 is assumed or used for correction or consideration. This appears to be a useful approximation, as fluctuations in the voltage division ratio with rotor position in a three-phase motor typically only range from 1% to 10% of the total amplitude. This approximation reduces the computational power required to account for the shunt voltage difference.

[0195] Figure 27 Shown as Figure 24An example of a comparison of the measured signal of an uncompensated or uncorrected induced voltage difference 3700 with half the value of the shunt voltage difference 3702 over the rotor rotation angle is shown. It can be seen that the halved shunt voltage difference 3702 contains a periodic component with half the electrical revolution period. If full compensation without approximation were used as described above, this periodic component would be omitted. However, according to some embodiments, the resulting residual error is not a problem for the reliability of the method used to determine the initial rotor position, so that the approximation with a fixed voltage division ratio can also produce consistently useful results.

[0196] In principle, there are three available intervals for north-south detection, from which the interval most suitable for the application can always be selected according to an optional embodiment, even in the case where the difference between the signal at the inductive voltage divider (i.e. the induced voltage difference) and the error signal (i.e. the shunt voltage difference) is maximum, so that the behavior of this difference around the zero crossing is irrelevant for determining the initial rotor position.

[0197] Figure 28 An example of a difference signal for north-south detection or for determining an initial rotor position is shown, corresponding to the difference between the induced voltage difference 3700 and the shunt voltage difference 3702. Figure 24 Compared to the (uncorrected) induced voltage difference signal shown in (which cannot be used to reliably determine the initial rotor position), Figure 28 The correction signal in has excellent symmetry and excellent signal quality, has reliable zero crossings, and is accordingly very suitable for reliably determining the initial rotor position.

[0198] The following explains, by way of example, the method for using the Figure 22A The control circuit determines the initial rotor position method.

[0199] A first phase and a second phase of a three-phase motor are initially energized with a current that increases or decreases during a first energizing interval.

[0200] In a further step, the induced voltage in the non-energized third phase of the three-phase motor is then determined at a first inductive measurement time point during the energization interval, and the voltage drop across the shunt resistor R1 of the control circuit is measured simultaneously or at the shortest possible time interval starting therefrom, which voltage drop is referred to as the shunt voltage drop.

[0201] After sufficient time has passed during the power-on interval to cause a current change (e.g., 1 ms), a second value of the induced voltage is measured at a second induction measurement time point in the first power-on interval, and the shunt voltage drop across the shunt resistor R1 is measured a second time simultaneously or at as short a time interval as possible therefrom.

[0202] The induced voltage difference is then determined by determining the difference between the first and second measured values ​​of the induced voltage and, to account for voltage variations in the control circuit, subtracting half the difference between the two measured shunt voltage drops (i.e., the halved shunt voltage difference). In this way, a corrected induced voltage difference is provided that is suitable for reliably determining the initial rotor position. Thus, the initial rotor position can be determined based on the determined induced voltage difference, at which voltage variations caused in the control circuit are corrected.

[0203] According to another alternative embodiment, the difference between the two shunt voltages multiplied by the actually measured inverse voltage divider ratio of the inductive voltage divider may be used instead of half the shunt voltage difference.

[0204] Another alternative method for determining the voltage drop in the control circuit to be compensated for can be to measure only the induced voltage or the induced voltage difference in a first energization interval, then reapply the same voltages in another energization interval and measure the shunt voltage difference in the second energization interval at the same measurement times as the induced voltage difference in the first energization interval. This makes it possible to determine the induced voltage difference and the shunt voltage difference using only one analog-to-digital converter.

[0205] Another optional possibility is to estimate the shunt voltage difference, as already explained above in the general description.

[0206] Another optional possibility for improving the signal can include using several energization intervals in which the applied voltage is alternately reversed and, for example, the determined voltage differences are summed and / or averaged. For this purpose, for example, the induced voltage difference and the shunt voltage difference can be added to each other and / or averaged.

[0207] The methods described herein are not limited to compensating for errors caused by voltage drops across one or more shunt resistors. Rather, these or similar methods can also be used to compensate for any other errors and / or asymmetries arising from the control circuit, for example, between the half paths following the positive and negative supply voltages. For this purpose, the induced errors can be measured and / or, if the asymmetry is known, calculated in a manner similar to the estimation described.

[0208] Reference Signs List

[0209] 10 Three-phase motor

[0210] 12 Phases of a three-phase motor

[0211] 12.1, 12.2, 12.3 The first, second and third phases of three-phase motors

[0212] 14 star points

[0213] 20 Control Unit

[0214] U1, U2, U3 Voltage at the first, second and third phase connections

[0215] L1, L2, L3 Inductance of the first, second and third phases

[0216] R1, R2, R3 Ohmic resistance of the first, second and third phases

[0217] i L1 、i L2 、i L3 Current of the first, second and third phases

[0218] U S Power supply voltage

[0219] g o,0 Initial switching upper limit of the switching interval

[0220] g u,0 Initial switching lower limit of the commutation interval

[0221] g o,i Adapt the switching upper limit of the reversing interval

[0222] g u,i Adapt the switching lower limit of the commutation interval

[0223] 100 Voltage difference curve measured without power

[0224] 200 Analog voltage difference curve without power

[0225] 300 Voltage difference curve measured when power is on

[0226] 400 Analog voltage difference curve when power is on

[0227] 500, 510, 700 commutation interval offset curve

[0228] 600, 610 switch limit distance curve

[0229] 210 three-phase motor

[0230] 212 Phases of a three-phase motor

[0231] 212.1, 212.2, 212.3 First, second and third phases of three-phase motors

[0232] 214 star points

[0233] U1, U2, U3 Voltage at the first, second and third phase connections

[0234] L1, L2, L3 Inductance of the first, second and third phases

[0235] R1, R2, R3 Ohmic resistance of the first, second and third phases

[0236] i L1 、i L2 、i L3 Current of the first, second and third phases

[0237] U S Power supply voltage

[0238] 2100 Voltage difference curve measured without power

[0239] 2200 analog voltage difference curve without power

[0240] 2300 Voltage difference curve measured when power is on

[0241] 2400 analog voltage difference curve when powered on

[0242] 310 three-phase motor

[0243] 312 Phases of a three-phase motor

[0244] 312.1, 312.2, 312.3 First, second, and third phases of three-phase motors

[0245] 314 star points

[0246] 320 control unit

[0247] U1, U2, U3 Voltage at the first, second and third phase connections

[0248] L1, L2, L3 Inductance of the first, second and third phases

[0249] R M1 、R M2 、R M3 Ohmic resistance of the first, second and third phases

[0250] i L1 、i L2 、i L3 Current of the first, second and third phases

[0251] U S Power supply voltage

[0252] M1 to M6 Field effect transistors for control circuits

[0253] R1(R2) shunt resistor

[0254] R2 to R5 are used for current measurement circuit

[0255] R6 to R 10 Ohmic resistance of the resistor divider

[0256] IC1 OPV for current measurement

[0257] 3100 Control Circuit

[0258] 3102 Control Circuit

[0259] 3700 Inductive voltage difference signal

[0260] 3702 Shunt voltage difference signal

Claims

1. A method for determining an initial rotor position of a three-phase motor (310) using a control circuit (3100), the method comprising: energizing a first phase (312.1) and a second phase (312.2) of a three-phase motor (310) with a current that increases or decreases during a first energizing interval; determining an induced voltage in a non-energized third phase (312.3) of a three-phase motor (310) at at least two temporally spaced measurement points in time during a first energization interval; determining an induced voltage difference (3700) of a voltage induced in a third phase (312.3) between two induction measurement time points spaced apart in time by taking into account a voltage change caused in a control circuit (3100) of the first phase (312.1) and / or the second phase (312.2) by energizing the first phase and the second phase in a first energization interval; and determining an initial rotor position based on the determined induced voltage difference (3700); It is characterized by taking into account the voltage change in the control circuit (3100) of the first phase (312.1) and / or the second phase (312.2) caused by the energization of the first phase and the second phase in the first energization interval in the form of a predetermined, estimated shunt voltage difference, or a voltage change in the control circuit of the first phase (312.1) and / or the second phase (312.2) caused by energizing the first phase (312.1) and / or the second phase (312.2) in a first energization interval corresponding to a shunt voltage difference (3702), which is generated at two temporally spaced shunt measurement time points at a voltage drop across a shunt resistor (R1) for measuring the phase current in the first phase (312.1) and / or the second phase (312.2), and In which, considering the energization of the first phase and the second phase in the first energization interval, the manner in which the voltage change is caused in the control circuit (3100) of the first phase (312.1) and / or the second phase (312.2) is the induced voltage difference (3700) minus a value proportional to the shunt voltage difference (3702).

2. The method according to claim 1, wherein The energization is performed such that the current increases or decreases strictly monotonically in the first energization interval.

3. The method according to claim 1, wherein The energization is performed such that the current increases or decreases strictly monotonically and linearly in a first energization interval.

4. The method according to any one of claims 1 to 3, wherein The sensing measurement time points include a first sensing measurement time point and a second sensing measurement time point, and the shunt measurement time points include a first shunt measurement time point and a second shunt measurement time point, wherein the first sensing measurement time point is no more than 5 μs away from the first shunt measurement time point, and / or the second sensing measurement time point is no more than 5 μs away from the second shunt measurement time point.

5. The method according to claim 4, wherein The first induction measurement time point and the second induction measurement time point are spaced apart from each other by at least 10 μs, and / or the first shunt current measurement time point and the second shunt current measurement time point are spaced apart from each other by at least 10 μs.

6. The method according to claim 4, wherein: The first induction measurement time point and the second induction measurement time point are spaced apart from each other by at least 100 μs, and / or the first shunt current measurement time point and the second shunt current measurement time point are spaced apart from each other by at least 100 μs.

7. The method according to claim 4, wherein: The first induction measurement time point and the second induction measurement time point are at least 1 ms apart from each other, and / or the first shunt measurement time point and the second shunt measurement time point are at least 1 ms apart from each other.

8. The method according to claim 1, wherein The voltage change in the first phase (312.1) and / or the second phase (312.2) caused by the energization of the first phase and / or the second phase in the first energization interval is considered to be the induced voltage difference (3700) minus half the shunt voltage difference (3702).

9. The method according to claim 1, further comprising: - energizing a first phase (312.1) and a second phase (312.2) of a three-phase motor (310) with a current that increases or decreases during a second energizing interval, wherein a voltage direction of the voltage applied for energizing is opposite to a voltage direction of the voltage used for energizing in the first energizing interval; - determining the induced voltage in the non-energized third phase of the three-phase motor (310) at at least two temporally spaced measurement time points during the second energization interval; - determining an induced voltage difference of the voltage induced in the third phase between two induction measurement time points spaced apart in time in the second energization interval by taking into account a voltage change in the control circuit of the first phase (312.1) and / or the second phase (312.2) caused by the energization of the first phase (312.1) and the second phase (312.2) in the second energization interval; - determining an initial rotor position based on the induced voltage difference determined in the first energization interval and the second energization interval.

10. The method according to claim 9, wherein: Before a first energizing interval and / or a second energizing interval, a voltage opposite to the voltage applied for energizing is applied to the first phase (312.1) and the second phase (312.2) for a predetermined period of time.

11. The method according to claim 10, wherein: The predetermined time period is no longer than 100 ms.

12. A control unit for a three-phase motor, characterized in that The control unit is configured to receive a phase voltage of each phase of the three-phase motor measured by a measurement and control circuit of the three-phase motor, and receive a voltage drop measured by a shunt resistor for measuring the current of each phase, and the control unit is configured to apply a maximum of three pulse voltages to the three-phase motor through the measurement and control circuit; as well as The initial rotor position of the three-phase motor is determined according to the method according to any one of claims 1 to 11 by using the voltage drop received by the measuring and control circuit and the phase voltage received by the measuring and control circuit and energizing each phase through the measuring and control circuit.

13. A system for determining an initial rotor position of a three-phase electric motor (310), the system comprising: measurement and control circuits; as well as The control unit according to claim 12; It is characterized by The measurement and control circuit is configured to measure the phase voltage of each phase of a three-phase motor (310) and the voltage drop across a shunt resistor for measuring the current of each phase, provide the measurement results to a control unit, and apply up to three pulse voltages to the three-phase motor.

14. An electric motor comprising a stator, a system for determining an initial rotor position of a three-phase electric motor (310) according to claim 13, and a rotor rotatable relative to the stator.

Images