Angular position error estimation at rest for high frequency voltage injection

Abstract

According to one aspect, a device for performing angular position error estimation is provided. The device uses a sensorless vector control method with high frequency voltage injection to operate (601) a synchronous motor (111) according to a predefined pulsating torque signal. The predefined pulsating torque signal has a zero mean and corresponds to a torque operating point. The device measures (602) a first value of an estimated angular position at a first time instance substantially corresponding to an endpoint of the predefined pulsating torque signal, and measures a second value of the estimated angular position and an estimated angular velocity at a second time instance occurring after the first time instance. Based on the measured values, the device estimates (603) an error in the estimated angular position and stores (604) the error in a lookup table. After performing the steps for one or more predefined pulsating torque signals, the device operates (607) the synchronous motor (111) using a lookup table for angular position error correction.

Description

Technical Field

[0001] Various example embodiments relate to controlling an industrial process. Background Art

[0002] In a synchronous motor, the rotation of the motor shaft (i.e., rotor) is synchronized with the frequency of the supply current in the steady state so that the rotation period is exactly equal to an integer number of alternating current (AC) cycles. A common way to control a synchronous motor is the so-called vector control, in which the torque of the motor is controlled by controlling the stator current vector. In order to implement such vector control, it is necessary to estimate the angular position of the rotor (e.g., using voltage injection). This estimation depends mainly on the saliency of the motor. One of the main challenges of any sensorless approach based on saliency is to overcome the errors introduced by the differential cross-coupling between the d-axis and the q-axis in the rotating reference frame.

[0003] US 10 224 851 B2 discloses a system and method for sensorless control of an electric machine using a magnetic alignment feature. WO 00 / 25418 A1 discloses a system and method for speed sensorless control of an induction motor. SILVERIO BOLOGNANI et al., "Sensorless Quasi-Stationary and Very Slow Speed ​​Position Detection in a Non-Shown PMSM Based on Current Injection and Back-EMF Observer" (SENSORLESS CONTROL FOR ELECTRICAL DRIVES (SLED), 2012 IEEE SYMPOSIUM ON, IEEE, September 21, 2012 (2012-09-21), pp. 1 to 7, P032322100, DOI: 10.1109 / SLED.2012.6422813, ISBN: 978-1-4673-2966-8) discloses a method for position detection of a permanent magnet synchronous motor based on detecting torque generation by signal injection. Specifically, two different application scenarios are disclosed: initial high-precision position detection at quasi-stationary, and position and speed tracking for closed-loop sensorless speed control. Summary of the invention

[0004] According to one aspect, the subject matter of the independent claims is provided. Embodiments are defined in the dependent claims.

[0005] One or more examples of implementations are set forth in more detail in the drawings and the description that follows. Other features will be apparent from the description and drawings, and from the claims.

[0006] Some embodiments provide apparatus, methods, and computer-readable media for performing angular position error estimation of a synchronous motor. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Hereinafter, example embodiments are described in more detail with reference to the accompanying drawings, in which

[0008] Figure 1 illustrates an exemplary industrial system according to one embodiment;

[0009] Figure 2 and Figure 3 Exemplary measurement results are illustrated;

[0010] Figure 4 illustrates an alternative exemplary industrial system according to one embodiment;

[0011] Figure 5 and Figure 6 illustrates a process according to an embodiment; and

[0012] Figure 7 An apparatus according to one embodiment is illustrated. DETAILED DESCRIPTION

[0013] The following embodiments are given as examples only. Although the specification may refer to "one", "an" or "some" embodiments and / or examples in several places in the text, this does not necessarily mean that one or more of the same embodiments or examples are referenced each time, or that a particular feature only applies to a single embodiment and / or example. Individual features of different embodiments and / or examples may also be combined to provide other embodiments and / or examples.

[0014] A synchronous motor is a motor in which the rotation of the motor shaft (i.e., rotor) is synchronized with the frequency of the supply current in a steady state so that the rotation period is equal to an integer number of alternating current (AC) cycles. A synchronous motor may include multi-phase AC electromagnets or permanent magnets on the stator of the motor that produce a magnetic field that rotates in phase with the oscillations of the line current. The rotor with permanent magnets or electromagnets rotates synchronously with the stator magnetic field at the same rate, thus providing a second synchronous rotating magnetic field of any AC motor. Some alternative synchronous motors (i.e., synchronous reluctance motors) generate torque through reluctance. In a synchronous reluctance motor, the stator consists of a plurality of protruding (protruding) electromagnet poles, while the rotor consists of a soft magnetic material with a plurality of protrusions that act as protruding magnetic poles through reluctance. A synchronous reluctance motor has the same number of stator poles and rotor poles.

[0015] A common way to control synchronous motors is the so-called vector control, in which the torque of the motor is controlled by controlling the stator current and / or voltage vector. The stator current / voltage vector is a vector comprising two orthogonal stator current / voltage components, which are derived from the stator current / voltage of a three-phase AC motor. One of the components defines the magnetic flux of the motor, while the other defines the torque of the motor. These stator current / voltage components can be written in a fixed reference frame as the α-axis stator current / voltage and the β-axis stator current / voltage, respectively, or written in a moving reference frame (moving synchronously with the rotor) as the d-axis stator current / voltage and the q-axis stator current / voltage. The actual control in vector control is usually performed for the d-axis stator current or voltage and the q-axis stator current or voltage, while the α-axis stator current or voltage and the β-axis stator current or voltage are used to actually drive the synchronous motor. The d-axis current / voltage and the q-axis current / voltage can be converted to the α-axis current / voltage and the β-axis current / voltage using an inverse Park transform, and the reverse conversion can be performed using a Park transform.

[0016] In order to perform the Park transform and the inverse Park transform, the angular position of the rotor (i.e., the rotation angle of the rotor) needs to be known. Voltage injection methods can be used to estimate the angular position of the rotor. This estimate depends on the saliency of the motor (i.e., the change in inductance at the motor terminals according to the rotor position). One of the main challenges in any saliency-based sensorless angular position estimation method is to overcome the errors caused by the differential cross-coupling between the d-axis and the q-axis (i.e., the direct axis and the quadrature axis) in the rotating reference frame. The d-axis and the q-axis are single-phase representations of the flux contributed by three independent sinusoidal phase quantities (traditionally represented by the letters "U", "V" and "W") at the same angular speed. The d-axis is the magnetic axis of the rotor, and the q-axis is the axis along which torque is generated (i.e., the torque can be directly controlled by adjusting the q-axis current). By convention, the quadrature axis always leads the direct axis by 90° in electrical angle, or in the more common case where the number of magnetic pole pairs can be greater than one, it leads by 180° / (2p), where p is the number of magnetic pole pairs. The cross coupling between the d-axis and the q-axis is usually described as an inductance in the voltage equation, as discussed below.

[0017] In the following, the problem of errors in angular position estimation in a synchronous motor controlled using vector control with voltage injection is discussed in a general sense and investigated analytically before a solution to the problem according to an embodiment is discussed in detail.

[0018] The following voltage equation describes the voltage balance in a non-hysteretic magnetic system where the only losses are related to metal losses (e.g., copper losses):

[0019]

[0020]

[0021] where u d and u q is the stator voltage associated with the d-axis and q-axis, i d and i q is the stator current associated with the d-axis and q-axis, R s is the resistance associated with resistive losses in the metal, λ d and λ q are the flux components of the d-axis and q-axis, and ω me is the angular velocity (or equivalently angular frequency) of the rotor. The flux linkage can be written as d =λ m +L d i d and λ q =L q i q , where λ m is the magnetic flux.

[0022] If the magnetic relationship between the d-axis current and the d-axis flux is also affected by the q-axis current (or vice versa when the q-axis quantity is affected by the d-axis current), then the inductive cross-coupling term L dq is introduced into the voltage equation. In this case, the voltage equation can be written (in matrix form):

[0023]

[0024] For high frequency injection on the d-axis, the following voltage waveform can be used:

[0025]

[0026] in is the amplitude of the d-axis stator voltage signal, ω h is the (high) angular frequency of the d-axis voltage signal, and t is time. Although in the above equations, high-frequency voltage injection is performed on the d-axis, it should be noted that high-frequency voltage injection can also be performed on the q-axis by defining the following equation:

[0027]

[0028] By applying the high frequency voltage injection in the previous equation to the voltage equation including the inductive cross-coupling term and further assuming that at relatively high frequencies (which means that R s is close to zero and can therefore be neglected) and at very low speeds (which means that ω me is close to zero and can therefore be ignored) Applying high frequency voltage injection, the voltage equation can be written as:

[0029]

[0030] Extracting the current from the voltage yields:

[0031]

[0032] Since the supplied injection voltage is given in the estimated dq reference frame, the estimated angular position θ is considered which is not necessarily equal to the actual angular position observer In the presence of an angular position error Δθ, the estimated angular position θ observer and the actual angular position θ me The relationship between observer =θ me +Δθ. The difference between the estimated dq reference frame and the actual dq reference frame can be written in the form of a rotation matrix:

[0033]

[0034] Therefore, the high-frequency current in the actual dq reference frame can be written as:

[0035]

[0036] Furthermore, can be demodulated (to remove the term sin(ω h t)) and is used to extract the rotor position:

[0037]

[0038] By setting The angular position error Δθ can be derived. The angular position error can be written as:

[0039]

[0040] In the presence of non-zero L dq In the case of , by setting Δθ = 0, the size of the error signal ε can be (based on the above The equation of is written as:

[0041]

[0042] Although in L dq In the special case where the angular position error is equal to zero, the error-free control to extract the rotor position, but in most practical cases this is not correct and compensation for the angular position error needs to be arranged in order to accurately estimate the angular position of the rotor. However, because the inductance L is directly measured dq , L dand L q This is not feasible, so it is not a simple task. The embodiments provide a solution that enables accurate estimation of the position error without a position sensor and (substantially) without a rotating synchronous motor.

[0043] Figure 1 A system for controlling a synchronous motor is illustrated. Specifically, Figure 1 The elements 101 to 110, 112, 113 for generating stator voltage signals for driving a synchronous motor 111, and a position estimation unit 124 for estimating the angular position of the rotor of the synchronous motor 111 are illustrated. As discussed above, the angular position of the rotor is required for driving the synchronous motor 111 in a precise manner using vector control, so the output of the position estimation unit 124 is used by the elements 101, 109, 113 driving the synchronous motor 111. The elements 101 to 110, 112 to 124 may form or be included in a drive (also referred to as a motor drive or electric drive). The synchronous motor 111 may be operated in industrial or non-industrial processes, for example, systems for transporting materials, such as pumps, fans, compressors, blowers, conveyors, cranes and / or elevators, and / or systems for processing materials, such as paper machines, mills, mixers and / or centrifuges. The illustrated system is based on (sensorless) vector control with high frequency voltage injection. Any of the elements 101 to 110 , 112 to 124 may be a logic element implemented as hardware or software or a combination thereof.

[0044] In the following, we start from the initial state where the system has not been fully deployed and the angular position error estimation has not yet been performed. Figure 1 The operation of the system after the unfolding process (ie, the angular position error estimation process) is discussed hereinafter.

[0045] Initially, the Δθ estimator state machine may maintain only information about one or more predefined pulsating torque signals in a database, wherein each predefined pulsating torque signal corresponds to a specific torque operating point (i.e., a specific torque level). The goal of the angular position error estimation process to be discussed below is to collect information about the error in the estimated angular position of the rotor of the synchronous motor 111 (as estimated by the illustrated system using high-frequency voltage injection) at one or more torque operating points by sequentially operating the synchronous motor 111 according to each of the one or more predefined pulsating torque signals. Preferably, the one or more torque operating points include a plurality of torque operating points and cover the entire operating range of the synchronous motor 111. The error in the estimated angular position is determined by the cross-coupling inductance L associated with the d-axis stator current and the q-axis stator current. dqinduced, as discussed above. Further, preferably, a plurality of predefined pulsating torque signals corresponding to a plurality of torque operating points are sequentially applied in an order of increasing torque operating points (i.e., starting from the lowest torque operating point and ending with the highest torque operating point). In other embodiments, another order (e.g., an order of decreasing torque operating points or a random order) may be employed. For all consecutive torque operating points, the intervals between consecutive torque operating points may be the same.

[0046] Each of the one or more predefined pulsating torque signals defined in the database of the Δθ estimator state machine 101 may include one or more positive pulses (e.g., positive square wave pulses) and one or more negative pulses (e.g., negative square wave pulses). Each of the one or more positive pulses preferably has the same peak amplitude. The peak amplitude of each predefined pulsating torque signal corresponds to the torque operating point of the synchronous motor. Further, each predefined pulsating torque signal has a zero mean so that the total movement of the rotor of the synchronous motor 111 is minimized when the synchronous motor 111 is driven according to the predefined pulsating torque signal. In other words, the initial angular position of the rotor before applying the predefined pulsating torque signal should be approximately the same as the final angular position of the rotor after applying the predefined pulsating torque signal. Preferably, the one or more positive pulses and the one or more negative pulses should also be arranged so as to minimize the total movement of the rotor caused by applying the predefined pulsating torque signal. In some embodiments, the predefined pulsating torque signal may, for example, be composed of a first duration t step A first positive (square wave) pulse having a second duration 2t step (i.e., twice the first duration) and a negative (square wave) pulse having a first duration t step The three pulses may be in the order described (or in some other order) and / or they may or may not follow each other directly (i.e., the time interval between consecutive pulses is zero or non-zero). All three pulses may have the same peak amplitude. Figure 2 One such example of a predefined pulsating torque signal is shown (discussed in detail below). To give another example, the predefined pulsating torque signal may be composed of a first negative (square wave) pulse, a negative (square wave) pulse, a second positive (square wave) pulse, and a second negative (square wave) pulse in sequence in other embodiments, wherein the duration of each pulse is the same. In this embodiment, any of the consecutive pulses may also directly follow each other or may not directly follow each other.

[0047] In order to estimate the error in the estimated angular position of the rotor for a given torque operating point, a sensorless vector control method with high frequency voltage injection is used (i.e., using elements 101 to 110, 112 to 124), and the synchronous motor 111 is operated according to a predefined pulsating torque signal corresponding to the torque operating point.

[0048] First, the predefined pulsating torque signal T generated by the Δθ estimator state machine 101 is calculated in the maximum torque per ampere (MTPA) element 102 to implement the MTPA control method. ref (n) is processed. The MTPA control method generates the desired torque while minimizing the current. The MTPA element 102 uses the MTPA control method based on the predefined pulsating torque pulse signal T ref (n) to generate the d-axis stator current signal and q-axis stator current signal In other embodiments, other devices may be used to convert the predefined pulsating torque signal into a d-axis stator current signal and a q-axis stator current signal.

[0049] Then, in element 103, a first difference signal is calculated as the generated d-axis stator current signal of the synchronous motor 111 The measured d-axis stator current signal i′ d Correspondingly, also in element 103, the second difference signal is calculated as the generated q-axis stator current signal of the synchronous motor 111 The measured q-axis stator current signal i′ q The first difference signal and the second difference signal quantify the error between the desired stator current and the actually measured stator current. How to actually measure the measured d-axis stator current signal i′ in the elements 112 and 113 will be described later. d and the q-axis stator current signal i′ q .

[0050] The d-axis stator voltage signal u is calculated based on the first difference signal by the first proportional integral (PI) controller 106. d Correspondingly, the q-axis stator voltage signal u is also calculated based on the second difference signal by the PI controller 106. q .

[0051] Optionally, to increase the dynamics of the control, in the combining (ie, summing) element 108, the d-axis stator voltage signal u d and q-axis stator voltage signal u qThe first selectable d-axis and q-axis stator voltage signal components may be combined (i.e., summed together) with the first selectable d-axis and q-axis stator voltage signal components, the second selectable d-axis and q-axis stator voltage signal components, or both, respectively. Specifically, the first selectable d-axis stator voltage signal component and the first selectable q-axis stator voltage signal component may be generated by the d-axis stator current signal and q-axis stator current signal The integral (I) controller 104 fed by the controller 104 calculates and the second optional d-axis voltage stator signal component and the second optional q-axis voltage stator signal component can be generated by the d-axis stator current signal and q-axis stator current signal The proportional (P) controller 105 fed is calculated. If the first optional d-axis voltage stator signal component and the first optional q-axis voltage stator signal component and the second optional d-axis voltage stator signal component and the second optional q-axis voltage stator signal component are calculated, the result signal is combined with the d-axis stator voltage signal u in the combination element 108. d and q-axis stator voltage signal u q Prior to combining, they may be combined in a combining (ie, summing) element 107. All operations discussed in this paragraph are performed on the d-axis signal and the q-axis signal separately (eg, d-axis signals are combined only with other d-axis signals, and q-axis signals are combined only with other q-axis signals).

[0052] In order to switch from the rotating reference frame defined by the d-axis and the q-axis to the stationary reference frame of the stator of the synchronous motor 111, the d-axis stator voltage signal u d and q-axis stator voltage signal u q The inverse Parker transform is used to transform the α-axis stator voltage signal u in the vector rotation (VR) element 109. α and β-axis stator voltage signal u β The inverse Parker transform requires knowledge of the current angular position (ie, rotation angle) of the rotor. As described in detail below, the current angular position may be estimated continuously during the running of the angular position error estimation process.

[0053] α-axis stator voltage signal u α and β-axis stator voltage signal u β and high frequency α-axis stator voltage signal and high frequency β-axis stator voltage signal are injected together into the combination element 110 (according to the high-frequency voltage injection principle) and a corresponding combination of these signals is fed to the stator of the synchronous motor 111.

[0054] The high frequency α-axis stator voltage signal and high frequency β-axis stator voltage signal As generated as follows. First, the signal generating element 121 in the position estimation unit 124 generates a high-frequency square wave d-axis stator voltage signal having a frequency greater than the frequency of the predefined pulsating torque signal. Preferably, the frequency of the high-frequency square wave d-axis stator voltage signal should be much greater than the frequency of the predefined pulsating torque signal (for example, one or more orders of magnitude or two or more orders of magnitude) so that the high-frequency component can be easily isolated from the main stator voltage / current signal into which it has been inserted by high-pass filtering. For example, the frequency of the high-frequency square wave d-axis stator voltage signal may be at least greater than the frequency of the predefined pulsating torque signal multiplied by ten or one hundred. Then, the high-frequency square wave d-axis stator voltage signal is transformed into the high-frequency α-axis stator voltage signal and the high-frequency β-axis stator voltage signal using an inverse Pike transform in the vector rotation element 122. Similar to what is discussed about the vector rotation element 109, at this point, it is assumed herein that the angular position of the rotor required to perform the inverse Pike transform is easily known. The following discusses in detail how to know the angular position (i.e., how to evaluate the angular position). As mentioned above, instead of the high frequency square wave d-axis stator voltage, a high frequency square wave q-axis stator voltage signal can also be used for the high frequency voltage injection.

[0055] The synchronous motor 111 generates an α-axis output stator current signal i that can be measured to generate an output α and β-axis output stator current signal i β These current signals are used for two different purposes. First, the α-axis output stator current signal i α and β-axis output stator current signal i β The vector rotation element 113 uses Park transformation to convert the d-axis output stator current signal i d and q-axis output stator current signal i q , the d-axis output stator current signal i d and q-axis output stator current signal i q After averaging in element 112, the stator current signals are output from the generated d-axis. and q-axis output stator current signal Specifically, the stator current signal i can be outputted for the d-axis in element 112. d and q-axis output stator current signal i q The purpose of calculating the average is to obtain the d-axis stator current signal i d and the q-axis stator current signal i qSimilar to what was discussed above with respect to elements 109 , 122 , the Park transform may be calculated in element 113 based on previously estimated (or predefined) values ​​of the angular position of the rotor maintained in a database of the Δθ estimator state machine 101 .

[0056] Next, in the position estimation unit 124, the α-axis output stator current signal i α and β-axis output stator current signal i β Used to estimate the rotor angular position and rotor angular velocity. In this case, the α-axis output stator current signal i α and β-axis output stator current signal i β In the vector rotation element 114, the Park transformation is first used to convert the q-axis output stator current signal i q Furthermore, in element 114, the Park transformation may be calculated based on a previously estimated or predefined value of the angular position of the rotor, as described in detail below. The stator current signal i is then output for the q-axis in elements 115, 116. q High-pass filtering is performed to isolate the high-frequency q-axis output stator current signal component Specifically, a unit delay is applied to the q-axis output stator current signal in the unit delay element 115, and the resulting delayed current signal is subtracted from the q-axis output stator current signal i in the subtraction element 116. q Here, the high frequency q-axis output stator current signal corresponds to the response of the synchronous motor 111 to the high frequency square wave d-axis stator voltage signal fed (or injected) into the synchronous motor 111, i.e., it is the signal component generated by the voltage injection.

[0057] The product of the high frequency q-axis output stator voltage component and the signal formed by applying a signal function to the high frequency square wave d-axis stator voltage signal in element 120 is calculated in multiplication element 117 .

[0058] Based on the product, the second PI controller 118 is used to calculate and estimate the angular velocity ω est Since this signal may be very noisy, at least in some embodiments, the signal is low-pass filtered in a low-pass filter (LPF) 119 to form a signal corresponding to the estimated angular velocity. A corresponding second signal. Specifically, the second signal The value of is used for angular position error estimation. The cutoff frequency of the LPF 119 can be defined based on Δt and the response time of the phase-locked loop (PLL) of the position estimation unit 124, where Δt represents the (predefined) time interval between a first measurement performed by the system at a first time instance while a predefined pulsating torque signal is being applied and a subsequent second measurement performed by the system at a second time instance when the predefined pulsating torque signal is no longer applied. Preferably, Δt should be as small as possible. Multiple values ​​of the (final) estimated angular velocity can be stored in a lookup table maintained in a database of the Δθ estimator state machine 101. In particular, the multiple values ​​of the (final) estimated angular velocity may include one or more of the following:

[0059] A first value of the estimated angular velocity measured at a first time instance corresponding to a non-zero torque of the predefined pulsating torque signal

[0060] a second value of the estimated angular velocity measured at a second time instance after the first time instance (specifically, occurring a predefined time Δt after the first time instance) and corresponding to zero torque in the predefined pulsatile torque signal as well as

[0061] The zeroth value of the estimated angular velocity measured at the zeroth time instance corresponding to zero torque and prior to the predefined pulsating torque signal and the first time instance (i.e. measured before the predefined pulsating torque signal is applied)

[0062] The first time instance may specifically (substantially) correspond to a time instance representing an end point of the predefined pulsating torque signal. In other words, the first time instance may be the last point having a non-zero torque value in the predefined pulsating torque signal. Since the predefined pulsating torque signal is defined so that the total movement of the rotor is minimized, the angular position of the rotor at the first time instance should be substantially the same as the angular position before the predefined pulsating torque signal is applied (see Figure 3 The third sub-graph from the top of FIG. 1 ). The zeroth time instance corresponds to the time instance before the predefined pulsating signal is applied (and thus also to the time instance before the first time instance and the second time instance). Here and hereinafter, the quantity and may specifically correspond to the low pass filtered quantity (as discussed with respect to element 119). Alternatively, the quantity and May correspond to an amount that has not been low-pass filtered (eg if a system according to an embodiment does not include the low-pass filter 119).

[0063] The estimated angular velocity ω is also calculated in the integrating element 123. est The first signal corresponding to the rotor is integrated (over time) to generate an estimated angular position θ of the rotor. est The latest value of this signal defines the current estimated angular position of the rotor used by the vector rotation elements 109, 113, 114, 122 during system operation. The value of the estimated angular position of the rotor can be stored in a lookup table maintained in the database of the Δθ estimator state machine 101. Specifically, the value of the estimated angular position of the rotor can include one or more of the following:

[0064] A second value of the estimated angular position measured at a second time instance (ie, after application of the predefined pulsating torque signal)

[0065] The zeroth value of the estimated angular position measured at the zeroth time instance

[0066] A first value of the estimated angular position measured at a first time instance

[0067] As discussed above, the zeroth time instance, the first time instance, and the second time instance are defined, that is, the measurements of the estimated angular position and the estimated angular velocity may be consistent with each other.

[0068] In addition to the aforementioned measurements of estimated angular position and estimated angular velocity, a consistent value of the desired torque (i.e., the value of the predefined pulsating torque signal) may also be measured and stored in a lookup table. In other words, one or more of the following values ​​may be measured and stored in a lookup table:

[0069] The zeroth value of the desired torque measured at the zeroth time instance (ideally a value of zero),

[0070] A first value of the estimated torque measured at a first time instance (corresponding to the torque operating point), and

[0071] A second value of the desired torque measured at a second time instance (Ideally a value of zero).

[0072] In some embodiments, the aforementioned torque values ​​(or at least some of them) may be pre-stored in a lookup table.

[0073] Figure 3 One example of measurements taken to calculate the error in the estimated angular position of the rotor is illustrated (discussed in detail later).

[0074] Based on the values ​​of the estimated angular position and the estimated angular velocity, the Δθ estimator state machine 101 calculates an estimate of the error in the estimated angular position for a torque operating point (at a first time instance associated with a non-zero torque) that defines a predefined pulsating torque signal. Specifically, the calculated error corresponds to the error in the first value of the estimated angular position. The calculated error is then stored in a lookup table in a database of the Δθ estimator state machine 101. Specifically, the calculation is performed as follows. As described above, during each Δθ estimation process for a single torque level (i.e., for a single torque operating point), the desired torque τ ref , estimated angular position θ est and the low-pass filtered estimated angular velocity Take measurements. Position error estimation requires at least the following three measurements:

[0075] a first value of an estimated angular position substantially corresponding to an endpoint of the predefined pulsating torque signal at a first time instance corresponding to a non-zero torque in the predefined pulsating torque signal;

[0076] A second value of the estimated angular position at a second time instance occurring a predefined time Δt after the first time instance (ie, after both the first time instance and the end of the predefined pulsating torque signal) as well as

[0077] A second value of the estimated angular velocity at a second time instance

[0078] The estimated angular position measured is a linear combination of the actual angular position φr and Δθ at the time of estimation. However, it can be assumed that at zero torque (τ ref =0), the error in the estimated angular position is negligible Δθ(0)≈0. Therefore, the three measurements of the estimated angular position are related to the actual angular position according to the following equation:

[0079]

[0080]

[0081]

[0082] From the intermediate equation shown above, the actual angular position must be known To calculate Δθ(τ ref). The actual angular position can be calculated using the estimated angular velocity at time t2 (low-pass filtered) and reintegrated over time Δt. Here, Δt is the duration between the first time instance and the second time instance. Assuming the term Δt is small enough, the following equation applies to the actual angular position at the first time instance:

[0083]

[0084] where the angular velocity of the rotor is assumed to be constant. Therefore, the error in the estimated angular position can be written as:

[0085]

[0086] It is worth noting that the calculation of the error in the estimated angular position described above employs only measurements made at the first time instance and the second time instance. The reliability of the process can be improved by utilizing measurements made at the zeroth time instance (i.e., before the predefined pulsating torque signal is applied). According to an embodiment, the initial estimated angular position (at the zeroth time instance) can be compared with each other. and the angular position at the first time instance To evaluate whether the system is operating in the expected manner. Ideally, if the predefined pulsating torque signal is applied correctly, the initial estimated angular position and the final estimated angular position should be equal.

[0087] During the angular position error estimation, the error in the estimated angular position is obviously not yet known. However, in order to accurately perform the Parker transform and the inverse Parker transform in elements 109, 113, 114, 122, and thus accurately perform the measurement and calculate the error, preferably, the error in the estimated angular position should be evaluated in some way. Such correction of the error in the estimated angular position during the angular position error estimation can, for example, be arranged as follows. Here, it can be assumed that one or more predefined pulsating torque signals are applied in increasing or decreasing order of torque operating points. For the nth torque operating point τ(n) among the one or more torque operating points, the corrected estimated angular position is calculated in the subtraction element 125 according to the following equation:

[0088]

[0089] where θ estis the value of the current estimated angular position as evaluated by the vector control method (specifically, by the position estimation unit 124), τ is the value of the torque associated with the torque operating point, Δθ(τ(n-1)) is the value of the error in the estimated angular position calculated for the previous torque operating point in the sequence (i.e., the value of the error in the previously calculated estimated angular position that most closely corresponds to the current torque operating point) and maintained in the lookup table, and x is defined as:

[0090]

[0091] For the first torque operating point in the sequence τ(1), a corrected estimated angular position may be calculated assuming a predefined value (e.g., zero) for the term Δθ(τ(0)). Preferably, the first torque operating point in the sequence τ(1) corresponds to the lowest torque operating point, such that Δθ(τ(0))≈0 is a reasonable assumption. The corrected estimated angular position is then used in the sensorless vector control method (specifically, for the inverse Park transform therein).

[0092] In some more general embodiments where predefined pulsating torque signals corresponding to different torque operating points may be applied in any order, the following equation may be used to calculate the corrected estimated angular position:

[0093]

[0094] Here, the term Δθ(τ(n-1)) has been replaced by the term Δθ(τ closest ), the term Δθ(τ closest ) is the previous torque operating point τ that is closest to the current torque operating point and maintained in the lookup table closest The value of the error in the estimated angular position is calculated. Therefore, the term τ closest Corresponding to indices less than or equal to n-1.

[0095] Even if the value of the estimated angular position is corrected during deployment by using the above equation, the estimated angular position will not be completely corrected. In order to further improve the correction, the value of the error in the previously calculated estimated angular position can be linearly extrapolated or interpolated to calculate the estimated angular position of τ(n) (i.e., the current torque operating point). For example, extrapolation can be used if one or more predefined pulsating torque signals are applied in the order of increasing or decreasing torque operating points. Interpolation can be used in the case where the value of the error in the estimated angular position to be calculated corresponds to a torque operating point located between two torque operating points where the error has been evaluated. In the subtraction element 125, the corrected estimated angular position is calculated for the nth torque operating point τ(n) among the one or more torque operating points according to the following equation:

[0096]

[0097] This option can be particularly beneficial for synchronous motors with linear Δθ, such as synchronous reluctance motors (SynRMs). However, for synchronous motors with nonlinear Δθ, such as interior permanent magnet (IPM) motors and permanent magnet assisted synchronous reluctance motors (PMaSynRMs), With φ r Significant errors may occur between the estimated and actual reference frames. Another option to further improve the correction is to reduce the torque step size (i.e., the interval between the nearest torque operating points or levels). Reducing the torque step size has many benefits, such as higher resolution of the torque vs. Δθ lookup table and smaller errors between the estimated reference frame and the actual reference frame.

[0098] The process discussed above for a single torque operating point (i.e., for a single predefined pulsating torque signal) can be repeated for multiple torque operating points (i.e., for multiple corresponding predefined pulsating torque signals). Preferably, the multiple torque operating points should cover the entire operating range of the synchronous motor and be evaluated sequentially in increasing order. The result of these repetitions is a lookup table comprising multiple values ​​for the error in the estimated angular position, wherein each of the multiple values ​​for the error corresponds to a specific torque operating point. The lookup table is then used during normal operation of the system for compensating for the error in the estimated angular position. Ideally, the torque level (i.e., torque operating point) is applied during the repetition in a pattern that returns the rotor to its starting position.

[0099] During normal operation (i.e., operation after the angular position error estimation phase in which the error is calculated and stored in a lookup table), Figure 1 The system can in most cases operate as described above for the angular position error estimation phase. Obviously, during normal operation, any torque signal (only opposite to the predefined pulsating torque signal) can be applied to the synchronous motor. Further, as mentioned above and described in detail below, during normal operation, the estimated angular position is corrected based on the lookup table (and the current torque operating point).

[0100] During normal operation (i.e., operation after the development or angular position error estimation phase in which the error is calculated and stored in the lookup table), for the nth torque operating point τ(n) of the one or more torque operating points, a corrected estimated angular position for the sensorless vector control method may be calculated in the subtraction element 125 according to the following equation:

[0101]

[0102] where θ estis the value of the current estimated angular position as evaluated by the vector control method, τ is the value of the torque associated with the torque operating point, Δθ(τ(n)) is the value of the error in the estimated angular position of the current torque operating point maintained in the lookup table (or specifically, the value of the error in the lookup table corresponding to the torque operating point that most closely matches the current torque operating point), and x is defined as:

[0103]

[0104] The corrected estimated angular position is then used in a sensorless vector control method (particularly for the inverse Parker transform therein).

[0105] In practice, a synchronous motor will not operate at a torque that corresponds exactly to any of the one or more predefined torque operating points at a given time. Therefore, in practice, the value of the error in the estimated angular position that is closest to the current torque operating point can be used during normal operation (i.e., operation after the angular position error is estimated) to estimate the error in the estimated angular position. In some embodiments, interpolation and / or extrapolation can be used to derive a more accurate estimate based on the values ​​in the lookup table (i.e., the value used to estimate the error in the torque operating point between the one or more torque operating points where data is stored in the lookup table).

[0106] Figure 2 shows a predefined pulsating torque signal τ according to an embodiment ref The illustrated predefined pulsating torque signal is composed of a first duration t step A first positive (square wave) pulse having a second duration 2t step A negative (square wave) pulse with a first duration t step The predefined pulsating torque signal shown corresponds specifically to a torque operating point (ie, torque operating level) of 70 Nm. In this example, t step = 0.01s. In general, t can be selected based on the bandwidth of the current controller and the minimum inertia of the rotor. step Ideally, the selected t step The value of is small enough to keep the rotor movement small, but large enough for the current controller to reach a stable state.

[0107] Figure 3 An example of how measurements can be performed in sequence to estimate the error in the estimated angular position for a single torque operating point is shown. Specifically, Figure 3 The first (top) subgraph of τ illustrates the desired torque τ ref The measurement of Figure 3 The second sub-figure illustrates the estimated angular position θest The measurement of Figure 3 The third sub-figure illustrates the actual angular position φ r The measurement of Figure 3 The fourth (bottom) sub-figure of illustrates the estimated angular velocity ω before and after low-pass filtering (LPF) r It should be noted that the measured angular position φ r exist Figure 3 According to any of the embodiments, it is not necessary to directly measure the actual angular position φ r The error Δθ in the estimated angular position can be calculated (eg, using a sensor). In the illustrated measurement, the cutoff frequency of the low pass filter (LPF) was set to 100 Hz, and according to Δt=4t step =0.04s, the term Δt is defined as a constant. Figure 3 The continuous measurement of the aforementioned quantities (not performed in the embodiment) and the instantaneous measurements made at the zeroth time instance (here about 2125.14s), the first time instance (here about 2125.18s) and the second time instance (here about 2125.22s) are illustrated, as defined above. In the fourth sub-figure, the hollow circles represent the unfiltered values ​​and the solid circles represent the low-pass filtered values. In the illustrated example, the following values ​​required for calculating the error in the estimated angular position are measured at the first time instance and the second time instance (indexes 1 and 2, respectively): and (The corresponding unfiltered value is ). The actual angular position values ​​provided for reference at the first and second time instances in the third sub-graph are and

[0108] Different types of synchronous motors use different dq conventions. Figure 1 The system is configured to align the d-axis along the path of minimum reluctance L'd>L'q, where L'd and L'q are the inductances along the d-axis and q-axis respectively. This results in a positive saliency ΔL'=L'd-L'q. This convention is applicable to many types of synchronous motors, such as SynRM and PMaSynRM. However, IPM motors rely more on permanent magnet (PM) torque than reluctance torque, and the typical convention is to align the d-axis along the PM flux path. This is usually the negative saliency path. In summary, Figure 1 The system cannot be used as such (without modification) with an IPM motor.

[0109] Figure 4 An alternative more general system according to an embodiment is illustrated for overcoming the aforementioned incompatibility with IPM motors. Figure 4 system and Figure 1 , except for one minor (but important) change illustrated by elements 401, 402, 403. That is, by enabling the system (or specifically, the Δθ estimator state machine 403) to switch the sign of the multiplication result in multiplication element 402 by including the correction term y in the multiplication, Figure 1 The system in Figure 4 The correction term y may have a value of +1 or -1 depending on the type of synchronous motor. Information about which sign is used at a given time may be maintained in a database of the Δθ estimator state machine 403. The Δθ estimator state machine 403 may control the switching of signs in the multiplication element 402. In other words, Figure 4 The following multiplication is performed in element 402:

[0110]

[0111] in

[0112] As implied above, the first type of synchronous motor includes at least SynRM and PMaSynRM, while the second type of synchronous motor includes at least IPM motor. In all other aspects, Figure 4 The system can be Figure 1 The system corresponds to that of Figure 4 The operation of the system is fully discussed.

[0113] As mentioned above, during the angular position error estimation process, the rotor of the synchronous motor should preferably move as little as possible. In other words, the initial angular position of the rotor before applying the predefined pulsating torque signal should be as close as possible to the final angular position of the rotor after applying the predefined pulsating torque signal. However, in some situations it may happen that there is a significant deviation between these two positions, and therefore, the accuracy of the error in the calculated estimated angular position is compromised.

[0114] Figure 5 A process for performing measurements on one or more torque operating points (preferably, multiple torque operating points) according to an embodiment is illustrated, thereby also overcoming the problems outlined in the previous paragraph. The process can be performed by Figure 1 or Figure 4 Specifically, the process can be performed by Figure 1 The Δθ estimator state machine 101 or Figure 4 The Δθ estimator state machine 403 is executed. In some embodiments, the process may be executed by a driver. Hereinafter, the entity executing the process is referred to as a device.

[0115] refer to Figure 5 In block 500, the apparatus first checks whether any of the one or more torque operating points remain for which an angular position error estimation has not yet been performed. If at least one torque operating point remains to be analyzed, then in block 501, the apparatus selects (in order) the next torque operating point τ for which an angular position error estimation is to be performed. ref . Initially, when no measurements have been performed, the actions in box 501 may simply correspond to selection of a first torque operating point among the one or more torque operating points. The first torque operating point among the one or more torque operating points may be the lowest torque operating point. If the one or more torque operating points include a plurality of torque operating points, a plurality of predefined pulsating torque signals corresponding to the plurality of torque operating points may be sequentially applied in an order of increasing torque operating points (i.e., increasing torque levels). In other embodiments, as discussed above, other orders may be employed.

[0116] Then, in block 502, the apparatus measures (at a zeroth time instance) a zeroth value of an estimated angular position of a rotor of a synchronous motor Specifically, the measurement is performed before any torque signal is applied to the synchronous motor. As described above, the measurement can be performed based on the high-frequency voltage injected into the stator of the synchronous motor. In addition, as described above, in some embodiments, the corresponding zeroth value of the expected torque and / or the estimated angular velocity of the rotor can also be measured.

[0117] After measuring the zeroth value(s), the apparatus applies a predefined pulsating torque corresponding to the selected torque operating point to the synchronous motor in block 503. This may be as described with respect to Figure 1 Performed as described.

[0118] In addition, with regard to Figure 1 Similar to the description, in block 504, the apparatus measures a first value of the estimated angular position at a first time instance during application of a predefined pulsating torque signal. Further, in block 505, the apparatus calculates a corrected value of the estimated angular position And further in block 505, the value is applied to the vector control (ie, at least to the Parker transform in the forward part of the vector control). Figure 1 The corrected value is calculated as described by the subtraction element 125 of . That is, the corrected value can be calculated for the initial value based on a predefined value (e.g., zero, assuming that the first torque operating point is relatively low) of the error in the estimated angular position. For (multiple) subsequent torque operating points, one of the aforementioned equations:

[0119] or

[0120]

[0121] can be employed (with extrapolation used to calculate Δθ(τ(n)) in the latter equation).

[0122] At a second time instance, in block 506, the device measures a second value of the estimated angular position and the second value of the estimated angular velocity As described above, the second time instance corresponds to a time instance after the predefined pulsating torque signal is applied. Based on the values ​​measured in boxes 505, 506, the device calculates the error Δθ (τ in the estimated angular position) in box 507. ref ) (or specifically, the error in the first value of the estimated angular position).

[0123] In block 508, the apparatus compares the absolute difference between the zeroth value of the estimated angular position and the error-corrected first value of the estimated angular position to a first predefined threshold (ie, a tolerance threshold). The error-corrected first value of the estimated angular position is defined as in is the second value of the estimated angular position, is a second value of the estimated angular velocity, and Δt is the duration between the first time instance and the second time instance. In other words, in block 508, the following equation is evaluated:

[0124]

[0125] where θ threshold is the first predefined threshold.

[0126] In response to the absolute difference not exceeding a first predefined threshold The device stores the error in the estimated angular position of the current torque operating point (and possibly information about the torque operating point and / or measurement data) to a lookup table in a database in block 511, and continues to perform angular position error estimation for the next torque operating point if a next torque operating point for which data has not been stored in the lookup table remains in the one or more torque operating points in block 500. In other words, the error estimation is considered successful, and the process continues.

[0127] In response to the absolute difference exceeding a first predefined threshold If the error estimation is considered to have failed (i.e., the error estimation is considered to have failed), the device may first check in block 509 whether the number of failures for which the absolute difference between the zeroth value of the estimated angular position and the first value of the estimated angular position exceeds the first predefined threshold value exceeds a second predefined threshold value. If this is not the case, the device repeats the angular position error estimation without changing the torque operating point (i.e., repeats at least blocks 502 to 508). If the number of failures exceeds the second predefined threshold value, the device may interrupt or stop the angular position error estimation process in block 510. The device may also interrupt the angular position error estimation process if no more torque operating points remain for angular position error estimation in block 500 (i.e., if all one or more torque operating points have been covered).

[0128] In some embodiments, the check in block 509 may be omitted. In some simpler embodiments, the tolerance check functions in blocks 502 and 508 may also be omitted, so that the Δθ(τ ref ) are all valid and therefore are stored in the lookup table in box 511.

[0129] exist Figure 5 After the angular position error estimation process is completed, the device can use the sensorless vector control method using high-frequency voltage injection to operate the synchronous motor at one or more torque operating points (or specifically within the range covered by one or more torque operating points). When performing the operation at one or more torque operating points, any error in the angular position of the synchronous motor can be corrected according to the lookup table.

[0130] Figure 6 A simplified process for performing angular position error estimation and controlling a synchronous motor according to the result of the angular position error estimation according to an embodiment is illustrated. The process can be Figure 1 or Figure 4 Specifically, the process can be performed by Figure 1 The Δθ estimator state machine 101 or Figure 4 The Δθ estimator state machine 403 of FIG. 4 is executed. In some embodiments, the process may be executed by a driver. Hereinafter, the entity executing the process is referred to as a device.

[0131] The apparatus performs angular position error estimation of the synchronous motor in the following manner (blocks 601 to 606) for each of the one or more torque operating points of the synchronous motor. Figure 5 As described, one or more torque operating points and associated one or more predefined pulsating torque signals may be defined as well as the order in which they are applied in the angular position error estimation.

[0132] In order to perform angular position error estimation for a specific torque operating point, in block 601, the apparatus operates a synchronous motor according to a predefined pulsating torque signal using a sensorless vector control method using high frequency voltage injection. Here, the peak amplitude of the predefined pulsating torque signal corresponds to the torque operating point of the synchronous motor, and the predefined pulsating torque signal has a zero mean value for minimizing the movement of the rotor of the synchronous motor during operation. The operation in block 601 may be, for example, Figure 5 In block 602, the apparatus measures a plurality of values ​​based on a high frequency component of an output stator current signal received from a synchronous motor (ie, during operation), the plurality of values ​​comprising at least:

[0133] a first value of the estimated angular position of the rotor of the synchronous motor at a first time instance corresponding substantially to an endpoint of the predefined pulsating torque signal (e.g., the last point of the predefined pulsating torque signal corresponding to a non-zero torque);

[0134] A second value of the estimated angular position of the rotor at a second time instance occurring a predefined time (Δt) after the first time instance as well as

[0135] A second value of the estimated angular velocity at a second time instance

[0136] The measurement in block 602 can be compared to Figure 5 The estimated angular velocity may correspond to the angular velocity measured after low pass filtering (e.g., Figure 1 as discussed in element 119).

[0137] In block 603, the apparatus estimates an error in a first value of an estimated angular position of the rotor based on the plurality of values, and stores the error in the estimated angular position at least in a lookup table in a database. Specifically, in block 603, the apparatus may estimate an error in a first value of an estimated angular position of the rotor based on the plurality of values, and stores the error in the estimated angular position in at least a lookup table in a database. The error (Δθ) in the first value of the estimated angular position is evaluated. The estimate in block 603 can be compared to Figure 5 Corresponding to box 507.

[0138] After storing the error in the lookup table in block 604, the device selects the next torque operating point for angular position error estimation (if any torque operating point exists in block 607) in block 608. The angular position error estimation loop (blocks 601 to 606) is repeated for each of the one or more torque operating points.

[0139] After performing angular position error estimation for all one or more torque operating points in blocks 601 to 606, the apparatus operates the synchronous motor at one or more torque operating points (or specifically, within a range covered by one or more torque operating points) using a sensorless vector control method with high frequency voltage injection in block 607. In operation, any error in the angular position of the synchronous motor is corrected according to the lookup table.

[0140] The above article uses Figure 1 , Figures 4 to 6 The described boxes, related functions and information exchanges are not in absolute time order, and some of them can be executed simultaneously or in an order different from the given order. Other functions can also be executed between or within them, and other information can be sent and / or other rules can be applied. Some boxes or parts of boxes or one or more pieces of information can be omitted or replaced by corresponding boxes or parts of boxes or one or more pieces of information.

[0141] Combination of the above Figures 1 to 5 Any of the features described in more detail in the embodiments may be combined with Figure 6 Combination of embodiments.

[0142] Figure 7 An apparatus 701 is provided, which is at least used to perform angular position error estimation for vector control of a synchronous motor. Specifically, Figure 7 A drive (or a subunit thereof) for performing an angular position error estimation for a vector control of a synchronous motor and performing the vector control itself (using high frequency voltage injection) based on the angular position error estimation may be illustrated. In some embodiments, the apparatus 701 may be used with a synchronous motor 111 that does not include the synchronous motor 111. Figure 1 The system corresponds to or does not include the synchronous motor 111 Figure 4 corresponding to the system.

[0143] The apparatus 701 may include one or more control circuit systems 720 (such as at least one processor) and at least one memory 730, the at least one memory 730 including one or more algorithms 731, such as computer program code (software), wherein the at least one memory and the computer program code (software) are configured to cause the apparatus to perform the above-described (i.e., Figure 1 or Figure 4 The at least one memory 730 may also include at least one database 732.

[0144] refer to Figure 7The one or more communication control circuit systems 720 include at least a vector control circuit system 721, which is configured to perform angular position error estimation and vector control of the synchronous motor taking into account the result of the angular position error estimation. To this end, the vector control circuit system 721 is configured to use one or more separate circuit systems by means of Figures 1 to 6 Specifically, the vector control circuit system 721 can implement Figure 1 Any of elements 101 to 110 and 112 to 125 and / or Figure 4 Any element of elements 102 to 110, 112 to 116, 118 to 125 and 401 to 403.

[0145] refer to Figure 7 , the memory 730 may be implemented using any suitable data storage technology, such as semiconductor-based memory devices, flash memory, magnetic memory devices and systems, optical memory devices and systems, fixed memory, and removable memory.

[0146] refer to Figure 7 , the device 701 may also include different interfaces 710 (such as an interface to the stator of a synchronous motor), and optionally one or more signaling interfaces (TX / RX) including hardware and / or software for implementing communication connectivity according to one or more communication protocols. The one or more signaling interfaces 710 may include standard well-known components such as amplifiers, filters, frequency converters, (de)modulators and encoder / decoder circuit systems controlled by corresponding control units, and one or more antennas.

[0147] As used in this application, the term "circuitry" may refer to one or more or all of the following: (a) a hardware-only circuit implementation, such as an implementation in analog and / or digital circuitry only; and (b) a combination of hardware circuitry and software (and / or firmware), such as (as applicable): (i) a combination of analog and / or digital hardware circuitry and software / firmware, and (ii) any portion of a hardware processor(s) with software, including a digital signal processor(s), software, and memory(s), which work together to enable an apparatus such as a terminal device or access node to perform various functions; and (c) hardware circuitry and processor(s) (such as a microprocessor(s) or a portion of a microprocessor(s)) that requires software (e.g., firmware) for operation, but may not be present when the software is not required for operation. The definition of "circuitry" applies to all uses of the term in this application, including any claims. As yet another example, as used in this application, the term "circuitry" also covers an implementation of only a hardware circuit or processor (or multiple processors) or a portion of a hardware circuit or processor and its (or their) accompanying software and / or firmware. For example and if applicable to a particular claim element, the term "circuitry" also covers a baseband integrated circuit for an access node or terminal device or other computing or network device.

[0148] In an embodiment, in combination Figures 1 to 6 At least some of the described processes may be performed by an apparatus including corresponding means for performing at least some of the described processes. Some example means for performing the processes may include at least one of the following: a detector, a processor (including dual-core processors and multi-core processors), a digital signal processor, a controller, a receiver, a transmitter, an encoder, a decoder, a memory, a RAM, a ROM, software, firmware, a display, a user interface, a display circuit system, a user interface circuit system, a user interface software, a display software, a circuit, an antenna, an antenna circuit system, and a circuit system. In an embodiment, at least one processor, a memory, and a computer program code form a processing device or include one or more computer program code portions for performing the processing according to the present invention. Figures 1 to 6 Any of the embodiments or operations thereof performs one or more operations.

[0149] The described embodiments may also be implemented in the form of a computer process defined by a computer program or part thereof. Figures 1 to 6Embodiments of the described methods can be performed by executing at least a portion of a computer program including corresponding instructions. The computer program can be provided as a computer-readable medium including program instructions stored thereon, or provided as a non-transient computer-readable medium including program instructions stored thereon. The computer program can be in source code form, object code form, or some intermediate form, and can be stored in a carrier, which can be any entity or device capable of carrying the program. For example, a computer program can be stored on a computer program distribution medium readable by a computer or processor. For example, a computer program medium can be, for example, but not limited to, a recording medium, a computer memory, a read-only memory, an electrical carrier signal, a telecommunication signal, and a software distribution packet. The computer program medium can be a non-transient medium. The coding of the software for executing the illustrated and described embodiments is completely within the scope of ordinary technicians in the art.

[0150] Even though the embodiments have been described above with reference to examples according to the accompanying drawings, it should be clear that the embodiments are not limited thereto but may be modified in several ways within the scope of the appended claims. Therefore, all words and expressions should be interpreted in a broad manner and they are intended to illustrate rather than restrict the embodiments.

Claims

1. A control device, comprising: Means for performing angular position error estimation, for performing said angular position error estimation for each of one or more torque operating points of a synchronous motor sequentially by: operating the synchronous motor according to a predefined pulsating torque signal using a sensorless vector control method with high frequency voltage injection, wherein a peak amplitude of the predefined pulsating torque signal corresponds to a torque operating point of the synchronous motor and a mean value of the predefined pulsating torque signal is zero; At least the following are measured based on the high frequency component of the output stator current signal received from the synchronous motor: A first value of the estimated angular position of the rotor of the synchronous motor at a first time instance corresponding to an end point of the predefined pulsating torque signal A second value of the estimated angular position at a second time instance occurring a predefined time Δt after the first time instance as well as a second value of the estimated angular velocity at the second time instance The error Δθ in the first value of the estimated angular position is estimated as as well as storing the error in the estimated angular position in at least a lookup table in a database; as well as Means for operating the synchronous motor, for operating the synchronous motor at the one or more torque operating points using the sensorless vector control method with high frequency voltage injection after the angular position error estimation, wherein in the operation at the one or more torque operating points, any error in the angular position of the synchronous motor is corrected according to the lookup table. 2 . The control device of claim 1 , wherein the one or more torque operating points comprises a plurality of torque operating points.

3. The control apparatus of claim 1 , wherein any error in the angular position of the synchronous motor is corrected in the operation at the one or more torque operating points after the angular position error estimation according to the lookup table by: The corrected estimated angular position for the torque operating point τ(n) is calculated according to the following equation: where θ est is the value of the current estimated angular position evaluated by the vector control method, τ is the value of the torque associated with the torque operating point, Δθ(τ(n)) is the value of the error in the estimated angular position maintained in the lookup table, n indicates the nth torque operating point of the one or more torque operating points, and x is defined as follows: as well as The corrected estimated angular position is used in the sensorless vector control method.

4. The control device according to claim 2, wherein any error in the angular position of the synchronous motor is corrected in the operation according to the predefined pulsating torque signal by: The corrected estimated angular position for the torque operating point τ(n) is calculated according to the following equation: where θ est is the value of the current estimated angular position evaluated by the vector control method, τ is the value of the torque associated with the torque operating point, Δθ(τ closest ) is the value of the error in the estimated angular position calculated for the previous torque operating point closest to the torque operating point and maintained in the lookup table, n indicates the nth torque operating point of the one or more torque operating points, and x is defined as follows: The corrected estimated angular position is used in the sensorless vector control method.

5. The control device according to claim 1, wherein the one or more torque operating points include at least three torque operating points, and any error in the angular position of the synchronous motor is corrected in the operation according to the predefined pulsating torque signal by: The corrected estimated angular position for the torque operating point τ(n) is calculated according to the following equation: where θ est is the value of the current estimated angular position evaluated by the vector control method, τ is the value of the torque associated with a torque operating point, Δθ(τ(n)) is the value of the error in the estimated angular position for the torque operating point τ(n), n indicates the nth torque operating point of the at least three torque operating points, and x is defined as follows: The term Δθ(τ(n)) is calculated by linearly extrapolating or interpolating the value of the error in the estimated angular position calculated for a previous torque operating point and maintained in the lookup table; and The corrected estimated angular position is used in the sensorless vector control method.

6. The control device according to claim 1, wherein the means for performing the angular position error estimation is configured to: perform the operation of the synchronous motor according to the predefined pulsating torque signal for each of one or more torque operating points of the synchronous motor by: generating a d-axis stator current signal and a q-axis stator current signal based on the predefined pulsating torque signal; calculating a difference between a generated d-axis stator current signal of the synchronous motor and a measured d-axis stator current signal as a first difference signal, and a difference between a generated q-axis stator current signal of the synchronous motor and a measured q-axis output stator current signal as a second difference signal; using a first proportional-integral (PI) controller to calculate a d-axis stator voltage signal based on the first difference signal and a q-axis stator voltage signal based on the second difference signal; Using an inverse Parker transform to convert the d-axis stator voltage signal and the q-axis stator voltage signal into an α-axis stator voltage signal and a β-axis stator voltage signal; as well as The α-axis stator voltage signal and the β-axis stator voltage signal are applied to a stator of the synchronous motor.

7. The control device according to claim 6, wherein the means for performing the angular position error estimation is configured to obtain the measured q-axis output stator current signal by: receiving an α-axis output stator current signal and a β-axis output stator current signal from the synchronous motor; generating a d-axis output stator current signal and a q-axis output stator current signal based on the α-axis output stator current signal and the β-axis output stator current signal using a Park transform based on a previous estimate of the estimated angular position of the rotor; and A time average over each of the d-axis output stator current signal and the q-axis output stator current signal is calculated.

8. The control device of claim 6, wherein the means for operating the synchronous motor at the one or more torque operating points is configured to perform the high frequency voltage injection by: Generate a high-frequency square wave d-axis stator voltage signal or a high-frequency square wave q-axis stator voltage signal having a frequency greater than the predefined frequency of the pulsating torque signal; converting the square wave d-axis stator voltage signal into a high frequency alpha-axis stator voltage signal and a high frequency beta-axis stator voltage signal using an inverse Parker transform based on a previous estimate of the estimated angular position of the rotor; and The high-frequency α-axis stator voltage signal and the high-frequency β-axis stator voltage signal are injected into the α-axis stator voltage signal and the β-axis stator voltage signal to be fed to the synchronous motor.

9. The control device according to claim 8, wherein the means for performing the angular position error estimation is configured to: perform the operation of the synchronous motor according to the predefined pulsating torque signal by performing the following for each of one or more torque operating points of the synchronous motor: receiving an α-axis output stator current signal and a β-axis output stator current signal from the synchronous motor; generating a q-axis output stator current signal based on the α-axis output stator current signal and the β-axis output stator current signal using a Park transform based on a previous estimate of the estimated angular position of the rotor; High-pass filtering the q-axis output stator current signal to isolate the high-frequency q-axis output stator current component generated by the high-frequency voltage injection; calculating the product of the high frequency q-axis output stator current component and a signal formed by applying a sign function to the high frequency square wave d-axis stator voltage signal; calculating, using a second PI controller, a first signal corresponding to the estimated angular velocity based on the product; integrating the first signal corresponding to the estimated angular velocity over time to form a signal corresponding to the estimated angular position; as well as The first signal corresponding to the estimated angular velocity is low pass filtered to form a second signal corresponding to the estimated angular velocity.

10. The control apparatus according to claim 8, wherein the means for performing the angular position error estimation is configured to: perform the operation of the synchronous motor according to the predefined pulsating torque signal using the sensorless vector control method using the high frequency voltage injection by performing the following for each of the one or more torque operating points of the synchronous motor: receiving an α-axis output stator current signal and a β-axis output stator current signal from the synchronous motor; generating a q-axis output stator current signal based on the α-axis output stator current signal and the β-axis output stator current signal using a Park transform based on a previous estimate of the estimated angular position of the rotor; high pass filtering the q-axis output stator current signal to isolate a high frequency q-axis output stator current component, wherein the high frequency q-axis output stator current component corresponds to a response to the high frequency square wave d-axis stator voltage signal fed to the synchronous motor; calculating the product of the high frequency q-axis output stator current component, a signal formed by applying a sign function to the high frequency square wave d-axis stator voltage signal, and a correction term maintained in the database, wherein the correction term has a value of +1 or -1 depending on the type of the synchronous motor; calculating, using a second PI controller, a first signal corresponding to the estimated angular velocity based on the product; integrating the first signal corresponding to the estimated angular velocity over time to form a signal corresponding to the estimated angular position; as well as The first signal corresponding to the estimated angular velocity is low pass filtered to form a second signal corresponding to the estimated angular velocity.

11. The control apparatus of claim 1 , wherein the one or more torque operating points comprises a plurality of torque operating points, and the means for performing the angular position error estimation is further configured to perform the following for each of the one or more torque operating points: measuring a zeroth value of said estimated angular position prior to said operation of said synchronous motor according to said predefined pulsating torque signal; The absolute difference between the zeroth value of the estimated angular position and the error-corrected first value of the estimated angular position is compared to a first predefined threshold, wherein the error-corrected first value of the estimated angular position is defined as In response to the absolute difference not exceeding the first predefined threshold, performing an operation of storing the error in the estimated angular position at least in a lookup table in a database, and if a next torque operating point remains in the one or more torque operating points for which data has not yet been stored in the lookup table, continuing to perform angular position error estimation for the next torque operating point; and In response to the absolute difference exceeding the first predefined threshold, the angular position error estimation is repeated without changing the torque operating point.

12. The control apparatus of claim 11, wherein the means for performing the angular position error estimation is further configured to perform the following for each of the one or more torque operating points: In response to a number of failures for which the absolute difference between the zeroth value of the estimated angular position and the first value of the estimated angular position exceeds the first predefined threshold exceeding a second predefined threshold, the angular position error estimation is stopped.

13. A control device according to any preceding claim, wherein the predefined pulsating torque signal comprises one or more positive square wave pulses and one or more negative square wave pulses, and the one or more positive square wave pulses and the one or more negative square wave pulses are arranged in an order that minimizes the total movement of the rotor during the application of the predefined pulsating torque signal.

14. A driver for driving a synchronous motor, the driver comprising a control device, the control device comprising: Means for performing angular position error estimation, for performing the angular position error estimation for each of the one or more torque operating points of the synchronous motor sequentially by: operating the synchronous motor according to a predefined pulsating torque signal using a sensorless vector control method with high frequency voltage injection, wherein a peak amplitude of the predefined pulsating torque signal corresponds to a torque operating point of the synchronous motor and a mean value of the predefined pulsating torque signal is zero; At least the following are measured based on the high frequency component of the output stator current signal received from the synchronous motor: A first value of the estimated angular position of the rotor of the synchronous motor at a first time instance corresponding to an end point of the predefined pulsating torque signal A second value of the estimated angular position at a second time instance occurring a predefined time Δt after the first time instance as well as a second value of the estimated angular velocity at the second time instance The error Δθ in the first value of the estimated angular position is estimated as as well as storing the error in the estimated angular position in at least a lookup table in a database; as well as Means for operating the synchronous motor, for operating the synchronous motor at the one or more torque operating points using the sensorless vector control method with high frequency voltage injection after the angular position error estimation, wherein according to the lookup table, in the operation at the one or more torque operating points, any error in the angular position of the synchronous motor is corrected.

15. A control method, comprising: The angular position error estimation is performed sequentially for each of the one or more torque operating points of the synchronous motor by: operating the synchronous motor according to a predefined pulsating torque signal using a sensorless vector control method with high frequency voltage injection, wherein a peak amplitude of the predefined pulsating torque signal corresponds to a torque operating point of the synchronous motor and a mean value of the predefined pulsating torque signal is zero; At least the following are measured based on the high frequency component of the output stator current signal received from the synchronous motor: A first value of the estimated angular position of the rotor of the synchronous motor at a first time instance corresponding to an end point of the predefined pulsating torque signal A second value of the estimated angular position at a second time instance occurring a predefined time Δt after the first time instance as well as a second value of the estimated angular velocity at the second time instance The error Δθ in the first value of the estimated angular position is estimated as as well as storing the error in the estimated angular position in at least a lookup table in a database; After the angular position error estimation, the method further comprises: The synchronous motor is operated at the one or more torque operating points using the sensorless vector control method with high frequency voltage injection, wherein any error in the angular position of the synchronous motor is corrected in the operation at the one or more torque operating points according to the lookup table.

Images