Device and method for determining the initial rotational position of a rotor of an electric motor that is at rest relative to a stator
By impressing positive and negative direct components on the current variable to form characteristic values, the method addresses angular errors in rotor positioning, achieving efficient and accurate rotor alignment with reduced losses and computation, suitable for electric motors.
Patent Information
- Application Number
- DE102023112992
- Authority / Receiving Office
- DE · DE
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-05-17
- Publication Date
- 2024-11-21
AI Technical Summary
Existing methods for determining the initial rotational position of a rotor relative to a stator in electric motors suffer from angular errors of 180°, leading to increased electrical losses and complex signal evaluation, particularly when using high-frequency injection signals.
A method that impresses both positive and negative direct components on the current variable to form characteristic values of alternating components, comparing these values to detect a potential 180° offset, thereby reducing electrical losses and computational effort.
The method allows for accurate determination of the initial rotational position with minimal electrical losses and reduced computational complexity, ensuring reliable rotor positioning without additional sensors.
Smart Images

Figure 00000000_0001_ABST 
Figure 00000000_0000_ABST
Abstract
Description
[0001] The invention relates to a method for determining an initial rotational position of a rotor of an electric motor that is stationary relative to a stator. Furthermore, the invention relates to a device for determining an initial rotational position of a rotor of an electric motor that is stationary relative to a stator.
[0002] The invention can be applied to field-oriented control of electric motors to determine the initial position of the electric motor's rotor. Typically, a so-called injection method is used to determine this initial rotational position, whereby the stator winding of the electric motor is excited with a high-frequency injection signal. The resulting current response of the electric motor is evaluated for position-dependent anisotropy in the magnetic structure of the electric machine in order to deduce the rotor's position. This procedure, as such, cannot distinguish between positive and negative orientation of the rotor's magnetic flux direction. Therefore, an angular error of 180° can occur when determining the initial rotational position.
[0003] A method for determining the initial rotational position of a rotor is known, for example, from EP 2 194 641 A1. In this method, a high-frequency injection signal is first applied, and an initial rotational position is estimated, which may contain an angular error of 180°. In a further step, the injection signal is modified to saturate the ferromagnetic material of the electric motor and subsequently detect any angular error of 180° that may be present. In this process, the amplitude of the current is drastically increased, which can lead to high losses and excessive heating of the electric motor. Furthermore, the evaluation of the response signal is comparatively complex.
[0004] Against this background, the task is to enable the determination of the initial rotational position with the lowest possible electrical losses and reduced computational effort.
[0005] The problem is solved by a method for determining an initial rotational position of a rotor of an electric motor that is at rest relative to a stator, wherein - in a first method step - a stator winding of the electric motor is excited with a high-frequency injection signal, a current variable of the electric motor influenced by the high-frequency injection signal is detected, and an initial rotational position of the rotor is estimated based on the detected current variable, wherein - in a second method step following the first method step - a positive DC component is impressed on the current variable while the current variable is detected and a first characteristic value, in particular a first effective value, of an alternating component of the current variable is formed, and a negative DC component is impressed on the current variable while the current variable is detected and a second characteristic value, in particular a second effective value, of an alternating component of the current variable is formed, wherein - in a third method step - the first characteristic value is compared with the second characteristic value in order to detect an offset of the initial rotational position of the rotor estimated in the first method step by 180°.
[0006] In the method according to the invention, to detect a possible offset of 180° – instead of increasing the amplitude of the current quantity as known from the prior art – a positive and a negative DC component are impressed on the current quantity. The two sub-steps of the second method step can be carried out one after the other in any order. Although the DC components lead to ohmic losses, they are significantly lower compared to the prior art. For evaluation, according to the invention, a characteristic value, for example the effective value, of the AC component of the current quantity is calculated, and the characteristic values thus determined are compared with each other. The computational effort required for this is also significantly lower than in the prior art.
[0007] According to an optional embodiment, the positive DC component and the negative DC component are identical in amount.
[0008] According to an advantageous embodiment, the first characteristic value is a sum or a discrete integral of the squared alternating component of the current variable, and the second characteristic value is a sum or a discrete integral of the squared alternating component of the current variable. By defining the characteristic value in this way, the required computational effort can be further reduced.
[0009] According to an advantageous embodiment, a 180° offset is detected if the second characteristic value is greater than the first characteristic value, and a 180° offset is excluded if the first characteristic value is greater than the second characteristic value. If the initial rotational position estimated in the first method step is correct, a positive DC component increases the saturation level of the iron. In this case, the inductance Ld in the rotor-related coordinate system is reduced, so that the amplitude of the measurable AC component of the current increases. Due to the increased current amplitude with a positive DC component, in this case (180° offset is excluded), the first characteristic value is greater than the second characteristic value determined with a negative DC component.However, if the initial rotational position estimated in the first step is rotated by 180°, the exact opposite situation arises: The positive DC component reduces the saturation level of the iron, the inductance Ld increases, and the amplitude of the measurable AC component of the current decreases. The lower current amplitude leads to a lower characteristic value than when the negative DC component is applied (180° offset detected).
[0010] According to an advantageous embodiment, a difference between the first and second characteristic values is calculated to compare the first characteristic value with the second characteristic value. The calculation of the difference can be determined with minimal computational effort.
[0011] In this context, it is advantageous to compare the magnitude of the resulting difference with a predefined minimum threshold. By comparing the difference with the minimum threshold, it can be determined whether the saturation of the ferromagnetic material of the electric motor has reached a required minimum value, so that the result of the offset determination can be trusted.
[0012] A particularly advantageous embodiment is one in which the second and third process steps are repeated if the magnitude of the resulting difference does not exceed the specified minimum threshold. In this case, the required minimum saturation is not reached. By repeating the process (and evaluating it), an attempt is then made to obtain a sufficient degree of saturation and thus a reliable offset determination result.
[0013] Preferably, when repeating the second method step, a positive DC component is used that is greater in magnitude than the positive DC component when the second method step was first performed, and / or when repeating the second method step, a negative DC component is used that is greater in magnitude than the negative DC component when the second method step was first performed. By increasing the magnitude of the DC components, increased saturation can be achieved during the repetition, thus increasing the probability of achieving a sufficient degree of saturation.
[0014] According to an advantageous embodiment, the second and third method steps are repeated until either the magnitude of the difference formed exceeds the specified minimum threshold or a specified number of repetitions is reached. The specified number of repetitions can serve as a termination criterion to ensure that the method is not carried out indefinitely if, for whatever reason, the required saturation level cannot be achieved.
[0015] According to an advantageous embodiment, the rotor is a permanent magnet rotor and / or the electric motor is a brushless DC motor. When the magnetic flux of the permanent magnet rotor saturates the iron core of the stator winding, a change in the alignment of the magnetic axes of the rotor relative to the toroidal winding leads to a change in the inductive properties of the winding. If the magnetic flux of the permanent magnet rotor is aligned with the winding, the iron core is maximally saturated and the coil inductance in the winding is reduced. This changes the current response depending on the rotational position of the rotor relative to the winding.
[0016] The electric motor can be arranged in a vehicle, in particular in an actuator or in a pump of the vehicle. The electric motor can have a winding with several ring windings arranged distributed around the circumference. The ring windings can have at least one ferromagnetic iron core.
[0017] The initial rotational position is preferably detected sensorlessly, i.e., without the use of a sensor measuring the rotational position of the rotor relative to the stator. This allows the electric motor to be constructed more cost-effectively. The cabling otherwise required for the sensor can be eliminated.
[0018] The initial rotational position is defined as the angular position of the rotor relative to the stator when the rotor is at rest, i.e., initially not rotating. The initial rotational position is preferably the angular position of the rotor before the electric motor generates drive power.
[0019] The electric motor can be arranged to operate at least one pump, in particular of the vehicle. The pump can provide a fluid pressure for actuating at least one vehicle component of the vehicle and / or a fluid volume flow for cooling at least one vehicle component of the vehicle.
[0020] A further subject of the invention is a device for determining an initial rotational position (20) of a rotor (12) of an electric motor (10) that is at rest relative to a stator (14), wherein the device is configured to - in a first method step - to excite a stator winding of the electric motor with a high-frequency injection signal, to detect a current variable influenced by the high-frequency injection signal and to estimate an initial rotational position of the rotor on the basis of the detected current variable, characterized in that the device is configured to - in a second method step following the first method step - impressing a positive DC component on the current variable while the current variable is being recorded and a first characteristic value, in particular a first effective value, of an AC component of the current variable is being formed, and impressing a negative DC component on the current variable while the current variable is being recorded and a second characteristic value, in particular a second effective value, of an AC component of the current variable is being formed, - in a third method step - to compare the first characteristic value with the second characteristic value in order to detect an offset of 180° of the initial rotational position of the rotor estimated in the first method step.
[0021] The device can achieve the same advantages and technical effects as have already been explained above in connection with the method according to the invention.
[0022] The advantageous embodiments and features described in connection with the method can also be used - alone or in combination - in the device according to the invention.
[0023] Further details and advantages of the invention will be explained below with reference to the exemplary embodiment illustrated in the drawings. Herein: Fig. 1 a schematic representation of an electric motor to explain an embodiment of the method according to the invention for detecting an initial rotational position of a rotor of the electric motor; Fig. 2 a flowchart of an embodiment of the method according to the invention; and Fig. 3 Time courses of the Id current according to a simulation of an embodiment of the method according to the invention.
[0024] Fig. 1 shows an electric motor 10 for implementing an embodiment of the method according to the invention. The electric motor 10 comprises a stator 14 and a rotor 12 rotatable relative to the stator 14. The rotor 12 is preferably a permanent magnet rotor 16, which is driven by an electrical control of a stator winding 18 of the stator 14. The stator winding 18 is reduced in size here, but in reality, the winding 18 is preferably distributed evenly over the circumference of the stator 14.
[0025] The stator winding 18 comprises at least one ferromagnetic iron core 19 and is electrically connected to a control unit, preferably an inverter, which adjusts electrical operating variables, in particular a voltage and / or a current, at the stator winding 18. The electrical operating variables are preferably related to the dq coordinate system of the rotor 12. The dq coordinate system is the reference system of the electrical operating variables that rotates with the rotor 12 and is obtained by the Park transformation.
[0026] When the rotor 12 is initially at rest relative to the stator 14, the initial rotational position 20, i.e., the rotational position of the rotor 12 relative to the stator 14, is crucial for the initial operation of the electric motor 10. In the present case, the initial rotational position 20 is detected sensorlessly, i.e., without additional sensors, by detecting electrical parameters of the electric motor 10, preferably by current response detection, in which a high-frequency injection signal 22 is introduced along a detection direction d in the dq coordinate system of the electric motor 10, and at least one current variable is detected as a current response to the injection signal 22 as the detection current variable I.
[0027] If a presumed d-direction d' is determined by applying current response detection in a first process step, a 180° offset may still exist between the presumed d-direction d' and the actual d-direction. This 180° offset can be detected using the method described in more detail below.
[0028] In a second method step following the first method step, a positive DC component is impressed on the current variable while the current variable is being recorded, and a first characteristic value, in particular a first effective value, of an AC component of the current variable is formed. Before or after this step, a negative DC component is impressed on the current variable while the current variable is being recorded, and a second characteristic value, in particular a second effective value, of an AC component of the current variable is formed.
[0029] In a third method step, the first characteristic value is compared with the second characteristic value in order to detect a possible offset of 180° of the initial rotational position of the rotor 12 estimated in the first method step.
[0030] An example procedure is shown below based on the illustration in Fig. 2. The method can be roughly divided into three method steps 101, 102, 103, wherein the first method step 101 proceeds as known in the prior art and is described, for example, in EP 2 194 641 A1. After switching on or activating 1000 the corresponding device and the electric motor, an RF injection method 1001 is carried out, in which the stator winding is excited with a high-frequency injection signal. The system waits 1002 until the tracking observer has stabilized. An estimated rotor angle is then adopted 1003 as the initial rotational position RtrIniPosn, and a counter is reset 1004 for the subsequent test procedure. This test procedure is carried out to detect and, if necessary, eliminate a possible offset of 180°.
[0031] As in Fig. 3a and Fig. As can be seen in Figure 3b, in the second method step 102, a positive DC component is first impressed on the current Id, which results in an increase by IdTest, followed by a negative DC component, which results in a reduction by IdTest. The impressed positive and negative DC components are identical in magnitude.
[0032] If the initial rotational position RtrIniPosn estimated in the first step is correct, the saturation level of the iron is increased by the positive DC component, see Fig. 3a. The inductance Ld or the impedance 2*pi * f_c * Ld decreases, so that the amplitude I_0+ of the alternating component of the current Id increases. The negative direct component, on the other hand, desaturates the iron. Therefore, the inductance increases due to the imposition of the negative direct component and reduces the amplitude I_0-.
[0033] The Fig. Figure 3b, however, shows the case where the initial rotational position RtrIniPosn estimated in the first step is misoriented by 180°. This arrangement has the opposite consequences: the positive DC component reduces the amplitude of the AC component of the current, and the negative DC component increases the amplitude of the AC component.
[0034] As further the presentation in Fig. 2, in the second method step 102, the DC component is first set to a predetermined value (see reference numeral 1005). In a subsequent sub-step 1006, the system waits until the current has stabilized. If the DC component is positive, a first characteristic value of an AC component of the current is calculated in a sub-step 1007, for example, over several periods of the fundamental oscillation. The polarity / sign of the DC component is then changed (see reference numeral 1008). Then, in sub-step 1009, a second characteristic value of an AC component of the current is calculated, for example, over several periods of the fundamental oscillation, and at the end of the second method step, the DC component is set to zero (see reference numeral 1010).
[0035] The use of the alternating component of the current Id does not require any additional computational effort compared to the state of the art. This alternating component is typically available anyway or is calculated using a bandpass filter as part of the injection method, see, for example, signal id_c in Fig. 1 of EP 2 194 641 A1.
[0036] In the third method step 103, the first characteristic value is compared with the second characteristic value in order to detect a 180° offset of the initial rotational position of the rotor 12 estimated in the first method step. To reduce the susceptibility of the method to interference, the effective values (or the root mean square values over the majority of the period of the carrier frequency f_c, RMS) of the alternating components of the current variable Id are preferably compared with one another as characteristic values. To further reduce the computational effort, the effective values can preferably be replaced by the sums or the discrete integrals of the squared alternating component of Id in the comparison.
[0037] In the third method step 103, the difference between the two characteristic values is first calculated (see reference numeral 1011). A plausibility check, or the verification of the degree of saturation for a "minimum saturation level," is performed by comparing the magnitude of the difference with a predefined minimum threshold. This can first be checked to see whether the difference is greater than a predefined positive minimum threshold (see reference numeral 1012), and then whether the difference is smaller than a predefined negative minimum threshold (see reference numeral 1013).
[0038] If the plausibility check fails, the value IdTest of the DC component is increased, for example, by 20%, and the second method step 102 is repeated. In this way, several attempts are made with increasingly higher offsets to induce pronounced saturation phenomena and thus achieve a plausible result. With each attempt, the counter for the subsequent test procedure is incremented (see reference numeral 1015). With each repetition, a sub-step 1014 checks whether the specified maximum number of attempts has not yet been reached. If this maximum number is exceeded without a plausible correction being possible, a warning is generated (see reference numeral 1016), and the device or electric motor enters an error state 1017, in which it is not operated.
[0039] If the aforementioned plausibility check is passed, a decision is made as to whether the correction is necessary. The correction is not necessary if the first characteristic value is greater than the second characteristic value, see reference numeral 1018. If, however, the first characteristic value is smaller than the second characteristic value, a correction must be made, see reference numeral 1019. After the corresponding correction or the corresponding update of the pole position 1020, the drive is enabled 1021 after a waiting period. The electric motor transitions to an operating state 1022 in which it is operated, in particular with a field-oriented control.
[0040] In the above-explained embodiment of a method for determining an initial rotational position 20 of a rotor 12 of an electric motor 10 that is at rest relative to a stator 14, - in a first method step - a stator winding of the electric motor 10 is excited with a high-frequency injection signal 22, a current variable of the electric motor 10 influenced by the high-frequency injection signal 22 is detected, and an initial rotational position of the rotor is estimated based on the detected current variable, wherein - in a second method step following the first method step - a positive DC component is impressed on the current variable while the current variable is detected and a first characteristic value, in particular a first effective value, of an alternating component of the current variable is formed, and a negative DC component is impressed on the current variable while the current variable is detected and a second characteristic value, in particular a second effective value, of an alternating component of the current variable is formed, - in a third method step - the first characteristic value is compared with the second characteristic value in order to detect an offset of the initial rotational position of the rotor 12 estimated in the first method step by 180°.
[0041] This allows the initial rotational position to be determined with the lowest possible electrical losses and reduced computational effort. QUOTES CONTAINED IN THE DESCRIPTION
[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature
[0000] EP 2 194 641 A1 [0003, 0030, 0035]
Claims
[1] Method for determining an initial rotational position (20) of a rotor (12) of an electric motor (10) which is at rest relative to a stator (14), wherein - in a first method step - a stator winding of the electric motor (10) is excited with a high-frequency injection signal (22), a current variable of the electric motor (10) influenced by the high-frequency injection signal (22) is detected, and an initial rotational position of the rotor is estimated on the basis of the detected current variable, characterized by , that - in a second method step following the first method step - a positive DC component is impressed on the current variable while the current variable is detected and a first characteristic value, in particular a first effective value, of an alternating component of the current variable is formed, and a negative DC component is impressed on the current variable while the current variable is detected and a second characteristic value, in particular a second effective value, of an alternating component of the current variable is formed, - in a third method step - the first characteristic value is compared with the second characteristic value in order to detect an offset of the initial rotational position of the rotor (12) estimated in the first method step by 180°. [2] Method according to claim 1, characterized by that the positive DC component and the negative DC component are identical in amount. [3] Method according to one of the preceding claims, characterized bythat the first characteristic value is a sum or a discrete integral of the squared alternating component of the first current and the second characteristic value is a sum or a discrete integral of the squared alternating component of the second current. [4] Method according to one of the preceding claims, characterized by that a 180° offset is detected if the second characteristic value is greater than the first characteristic value and a 180° offset is excluded if the first characteristic value is greater than the second characteristic value. [5] Method according to one of the preceding claims, characterized by that in order to compare the first characteristic value with the second characteristic value, a difference is formed between the first and the second characteristic value. [6] Method according to claim 5, characterized by that an amount of the difference formed is compared with a predetermined minimum threshold. [7] Method according to claim 6, characterized bythat the second and third procedural steps are repeated if the amount of the difference formed does not exceed the specified minimum threshold. [8] Method according to claim 7, characterized by that when the second method step is repeated, a positive DC component is used which is greater in magnitude than the positive DC component when the second method step was carried out for the first time and / or that when the second method step is repeated, a negative DC component is used which is greater in magnitude than the negative DC component when the second method step was carried out for the first time. [9] Method according to claim 7 or 8, characterized by that the second and third process steps are repeated until either the amount of the difference formed exceeds the specified minimum threshold or a specified number of repetitions is reached. [10] Device for determining an initial rotational position (20) of a rotor (12) of an electric motor (10) that is at rest relative to a stator (14), the device being configured to - in a first method step - to excite a stator winding of the electric motor (10) with a high-frequency injection signal (22), to detect a current variable influenced by the high-frequency injection signal (22) and to estimate an initial rotational position of the rotor on the basis of the detected current variable, characterized by that the device is configured to - in a second method step following the first method step - impressing a positive DC component on the current variable while the current variable is being recorded and a first characteristic value, in particular a first effective value, of an AC component of the current variable is being formed, and impressing a negative DC component on the current variable while the current variable is being recorded and a second characteristic value, in particular a second effective value, of an AC component of the current variable is being formed, - in a third method step - to compare the first characteristic value with the second characteristic value in order to detect an offset of the initial rotational position of the rotor (12) estimated in the first method step by 180°.
Citation Information
Patent Citations
System for recording the initial pollage of an electromotor runner
EP2194641A1
