Sensorless control method, variable speed drive, assembly and storage medium

Abstract

A method for sensorless control of an electrical motor (14) implemented in a variable speed drive (12) is presented, comprising: determining a control voltage (u0) to be applied to the motor (14); injecting a high frequency signal (S0) to the control voltage (u0) to obtain an excitation voltage (u1), wherein one or more frequencies of the high frequency signal (S0) vary over time; applying the excitation voltage (u1) to the motor (14); measuring a current (y) induced in the motor by the excitation voltage (u1), wherein the current (y) comprises a fundamental current (y a ) induced by the control voltage (u0) and a disturbance current (y b S1) induced by the high frequency signal (S0); and demodulating the current signal (y).

Description

Technical Field

[0001] This disclosure relates to a control method for an electric motor. This disclosure also relates to a speed-changing drive capable of controlling an electric motor. This disclosure further relates to an electrical drive assembly including a speed-changing drive and an electric motor. Background Technology

[0002] Electric motors (such as induction motors or permanent magnet synchronous motors (PMSMs)) are typically controlled by variable speed drives (VSDs) that use vector control. Vector control, or field-oriented control (FOC), uses two orthogonal components to determine the voltage to be sent to each motor stator winding. One component defines the magnetic flux generated by the stator, while the other corresponds to the torque determined by the motor's speed.

[0003] FOC (Focus of Control) relies on the accurate acquisition of the motor's rotor position. Rotor position is typically obtained through optical or magnetic transducers (encoders) or additional windings in the rotor (resolvers). However, this introduces additional complexity into the system, increases manufacturing costs, and reduces reliability. Furthermore, the use of optical encoders can be a source of measurement error, which may degrade FOC performance.

[0004] Solving these problems has led to the development of sensorless control strategies. One type of sensorless control strategy is high-frequency (HF) signal injection. This strategy involves injecting a high-frequency signal superimposed on the voltage of the motor and extracting the rotor position from the high-frequency current induced by the injected signal.

[0005] High-frequency signal injection has shown promising results compared to other sensorless control strategies, especially at low motor speeds. However, it has also been shown to generate acoustic noise, vibration, and additional losses in the motor. Reducing the amplitude of the high-frequency signal can reduce acoustic noise, but it also reduces the signal-to-noise ratio (SNR), making it more difficult to extract the information contained in the high-frequency current.

[0006] The present invention aims to provide a high-frequency signal injection control strategy that does not have the above-mentioned drawbacks. Summary of the Invention

[0007] A method for sensorless control of an electric motor implemented in a variable speed drive is proposed, wherein the method includes: Determine the control voltage to be applied to the motor; A high-frequency signal is injected into the control voltage to obtain the excitation voltage, wherein one or more frequencies of the high-frequency signal vary with time; Apply excitation voltage to the motor; The current induced in the motor by the excitation voltage is measured, wherein the current includes the fundamental current induced by the control voltage and the interference current induced by the high-frequency signal; Demodulated current signal, wherein the demodulated current signal includes: A first finite impulse response (FIR) filter is applied to the current signal to extract the fundamental current, wherein the duration of the response from the first FIR filter varies according to one or more frequencies of the high-frequency signal, and The interference current is extracted by subtracting the fundamental current from the current signal. Determine the amplitude of the interference current ripple, which depends on the state of the motor.

[0008] Therefore, when an excitation voltage is applied to the motor, one or more variable frequencies of the high-frequency signal can expand the spectrum of the acoustic noise generated by the motor. Furthermore, the demodulated current signal is adapted to one or more varying frequencies of the high-frequency signal to obtain motor information.

[0009] The following features can be implemented, either individually or in combination with other features: Determining the amplitude of the ripple involves calculating the zero-mean primitive of the high-frequency signal and multiplying the interference current by the zero-mean primitive. Determining the amplitude of the ripple also includes applying a second finite impulse response filter, wherein the duration of the response from the second finite impulse response filter varies according to one or more frequencies of the high-frequency signal; One or more frequencies of a high-frequency signal change randomly over time; High-frequency signals are sinusoidal signals; High-frequency signals are square wave signals; Determining the control voltage includes using a current controller; applying the excitation voltage to the motor includes using a pulse width modulation carrier to convert the excitation voltage into a pulse width modulation voltage; and the fundamental frequency of the high-frequency signal is within the range defined by the frequency bandwidth of the current controller and the frequency of the pulse width modulation carrier. The fundamental frequency of high-frequency signals varies between 250 Hz and 1 kHz; The control voltage is determined in the estimated rotor reference frame of the motor, and a high-frequency signal is injected into the estimated rotor reference frame. The control method also includes forming a feedback loop by updating the control voltage with the fundamental current and the amplitude of the ripple.

[0010] On the other hand, a variable speed drive for an electric motor is proposed, which includes a processor and a memory, the processor being configured to operate according to any of the methods described above.

[0011] On the other hand, an electric drive assembly is proposed, which includes a variable speed drive and an electric motor controlled by the variable speed drive.

[0012] In another aspect, a computer-readable storage medium comprising instructions that, when executed by a processor, cause the processor to perform the methods described above. Attached Figure Description

[0013] Other features, details, and advantages will be shown in the following detailed description and accompanying drawings, wherein: Figure 1 An example of a variable speed drive connected to the power grid and a motor is shown schematically.

[0014] Figure 2 The schematic diagram illustrates the methods used to implement the operation. Figure 1 An example of a control system for a motor control method.

[0015] Figure 3 schematically shown Figure 2 Details of the control system.

[0016] Figure 4 schematically shown Figure 2 Another detail of the control system.

[0017] Figure 5 It shows in Figure 2 Examples of signals used in a control system. Detailed Implementation

[0018] Figure 1 This is a schematic diagram of the electrical drive assembly 10 according to the present disclosure. The electrical drive assembly 10 can be used in various industrial settings. For example, it can drive fans in heating, ventilation, and air conditioning (HVAC) systems. As another example, it can also be used to drive water pumps in sewage facilities. As yet another example, it can drive conveyors on assembly lines. Many other industrial applications can be envisioned by those skilled in the art.

[0019] The electric drive assembly 10 includes a variable speed drive 12 (or VSD) and an electric motor 14.

[0020] Preferably, the electric motor 14 is an AC motor 14, and more preferably a synchronous motor, such as a permanent magnet synchronous motor (PMSM) or a synchronous reluctance motor (SynRM).

[0021] The speed changer 12 is electrically connected to the electric motor 14. The speed changer 12 controls the operation of the electric motor 14 according to a control method. The speed changer 12 enables the electric motor 14 to operate at the speed required by the application. The speed changer 12 also allows control over the torque output of the electric motor 14 to the load.

[0022] The variable speed drive 12 generally includes a rectifier module 16, a DC power bus 18, and an inverter module 20.

[0023] The rectifier module 16 includes a diode bridge configured to convert the three-phase AC voltage supplied by the power grid G ​​into a DC voltage. The DC voltage output from the rectifier module 12 can be applied to the DC power bus 18.

[0024] DC power bus 18 includes two power lines connected together by a bus capacitor Cbus, which is configured to stabilize the voltage of bus 18. The output of DC power bus 18 can be connected to inverter module 20.

[0025] Inverter module 20 includes multiple switching arms, each including a power transistor, such as an IGBT (Insulated Gate Bipolar Transistor) type power transistor. Inverter module 20 can be designed to cut off the voltage supplied by DC power bus 16 to achieve a variable output voltage that can operate electric motor 14.

[0026] In addition, the transmission drive 12 includes a processor PROC. The processor PROC controls other electrical components of the transmission drive 12. The processor PROC is configured to operate according to a control method. The processor PROC may include electronic circuitry for calculations managed by the operating system.

[0027] The variable speed drive 12 also includes a non-transitory machine-readable or computer-readable storage medium, such as a memory or storage device unit MEM, wherein the non-transitory machine-readable storage medium is encoded with instructions executable by a processor PROC, the machine-readable storage medium including instructions for operating the processor PROC to perform according to a control method. The computer-readable storage device according to this disclosure can be any electronic, magnetic, optical, or other physical storage device storing executable instructions. The computer-readable storage device can be, for example, random access memory (RAM), electrically erasable programmable read-only memory (EEPROM), a storage drive, and an optical disk, etc.

[0028] Figure 2 A control system for controlling the operation of motor 14 by the speed drive 12 is shown. The control system enables the speed drive 12 to operate according to the control method described below.

[0029] The control method includes, in step 101, determining a control voltage u0. The control voltage u0 is a voltage that, when applied to the motor 14, will cause the motor to operate under the conditions required for the application. For example, the control voltage u0 may enable the motor 14 to operate at a desired speed.

[0030] Here, in step 101, it is determined that the control voltage u0 of the motor 14 is achieved through field-oriented control or FOC.

[0031] Figure 3Step 101 for determining the control voltage u0 using FOC is shown in more detail.

[0032] As shown in the figure, determining the control voltage u0 depends on the reference speed ωref, the flux reference Φref, and information about the state of motor 14. The speed reference ωref is the speed at which motor 14 is expected to operate for the application. The flux reference Φref can be derived from the speed reference ωref and the motor specifications. Here, the information about the state of motor 14 includes the flux Φ generated by motor 14 and the actual speed ω.

[0033] Information about the motor's state is derived from measurements taken at motor 14. The control system is a closed-loop system. In other words, determining the control voltage u0 is achieved by receiving feedback about the instantaneous state of motor 14 during its operation. The transmission drive 12 adjusts the control voltage u0 based on the received feedback.

[0034] Here, the measurements performed at the motor are "sensorless." The control method is sensorless. This means that the feedback relies entirely on the current measurement provided by the current sensor embedded in the transmission drive 12. There are no external sensors, such as shaft encoders, mounted on the motor 14 to provide feedback to the transmission drive 12 about the state of the motor 14.

[0035] like Figure 3 As shown, determining the control voltage u0 of motor 14 includes using a speed controller and a flux controller. In step 1010, the speed controller derives the torque-generating current Iq from a reference speed ωref and the actual speed ω of motor 14. In step 1011, the flux controller derives the magnetizing current Id from a flux reference Φref and the actual flux Φ of motor 14. The speed controller and flux controller can use proportional-integral (PI) control to derive the torque-generating current and the magnetizing current. The torque-generating current Iq and the magnetizing current Id are two orthogonal signals.

[0036] Step 1012, determining the control voltage u0, also includes using a current controller. The current controller converts the two quadrature signals Iq and Id into the control voltage u0. The current controller can use current measurements taken at motor 14 to convert the two quadrature signals Iq and Id. It should be noted that the control voltage u0 determined according to FOC also includes two quadrature components (not shown here for simplicity).

[0037] Determining the control voltage u0 also includes, in step 1013, changing the reference frame of the control voltage u0. In practice, the control voltage u0 can be expressed in either a rotor reference frame or a stator reference frame. Typically, the control voltage u0 is initially determined in an estimate in the rotor reference frame and then converted to the stator reference frame. The control voltage u0 expressed in the stator reference frame defines the voltage to be applied to the stator windings to obtain the desired speed or torque from the motor 14.

[0038] return Figure 2 The control method includes, in step 102, injecting a high-frequency signal S0 into the control voltage u0 to obtain an excitation voltage u1. The excitation voltage u1 is a combination of the control voltage u0 and the high-frequency signal S0. In other words, the high-frequency signal S0 integrates the control voltage u0.

[0039] The high-frequency signal S0 can be a signal comprising multiple frequencies, particularly the fundamental frequency and any number of harmonic frequencies. For example, the high-frequency signal S0 can be a square wave signal. The use of square wave signals simplifies the synthesis of the high-frequency signal S0. Alternatively, the high-frequency signal S0 can be a sine wave signal or a signal of any other shape.

[0040] One or more frequencies of the high-frequency signal S0 vary over time. These frequencies vary randomly over time. In other words, one or more frequencies of the high-frequency signal S0 do not follow a repetitive pattern. When the excitation voltage u1 is applied to the motor 14, the random variation of the high-frequency signal S0 expands the spectrum of the acoustic noise generated in the motor 14. This can reduce the audible noise generated in the motor 14.

[0041] One or more frequencies of the high-frequency signal S0 are higher than the frequency bandwidth of the current controller used to determine the control voltage u0 in step 101. Therefore, the high-frequency signal S0 does not interfere with the desired operation of the motor 14. For example, the fundamental frequency of the high-frequency signal S0 can vary between 250 Hz and 1 kHz.

[0042] In the example shown, a high-frequency signal S0 is injected into the estimated rotor reference frame. Alternatively, a high-frequency signal S0 can be injected into the stator reference frame. Furthermore, a high-frequency signal S0 can be injected into any quadrature component that forms the control voltage u0.

[0043] The control method includes, in block 103, converting the excitation voltage u1 into a pulse width modulation voltage upwm. The conversion of the excitation voltage u1 into the pulse width modulation voltage upwm uses a pulse width modulation carrier ucar. Using a pulse width modulation carrier ucar means modulating the excitation voltage u1 with the pulse width modulation carrier ucar.

[0044] The frequency of the pulse-width modulated carrier ucar is higher than the fundamental frequency of the high-frequency signal S0. For example, the frequency of the pulse-width modulated carrier ucar can be between 2 and 16 kHz. Therefore, the fundamental frequency of the high-frequency signal S0 is within the range defined by the frequency bandwidth of the current controller and the frequency of the pulse-width modulated carrier ucar. Preferably, the fundamental frequency of the high-frequency signal S0 is close to the frequency of the pulse-width modulated carrier ucar to avoid interfering with the desired operation of the motor 14.

[0045] It should be noted that Figure 2 A single-phase control system 22 is shown. This is for simplicity only. Typically, the electric motor 14 will be a three-phase motor. In this case, in step 103, the excitation voltage u1 is converted to generate a pulse width modulation voltage upwm for each of the three phases of the motor 14.

[0046] The control method includes, in step 104, applying an excitation voltage u1 to the motor 14. Here, the excitation voltage u1, in the form of a pulse width modulated voltage upwm, is sent to the converter 20 of the speed drive 12 to operate the motor 14. The motor 14 will respond to the excitation voltage u1 by rotating at a speed ω and generating a current y in the stator windings.

[0047] The control method includes, in step 106, measuring the current y induced by the motor 14 during operation. As described above, the current measurement is provided by the transmission drive 12. The current y measured in step 106 includes the fundamental current ya and the interference current yb S1.

[0048] The fundamental current ya is the fundamental current induced by the control voltage u0. The fundamental current ya will be measured at step 102 without injecting a high-frequency signal S0. The fundamental current ya can be fed back to control step 101 to update the control voltage u0. Specifically, the current controller can use the fundamental current ya.

[0049] The interference current yb S1 is the interference current induced by the high-frequency signal S0. The interference current yb S1 is the primitive S1 of the injected high-frequency signal S0, whose amplitude is modulated by the ripple yb. The amplitude of the ripple yb is the amplitude of the motor's response to the high-frequency signal S0 and contains information about the motor's state used in step 101 to determine the control voltage u0. Therefore, the amplitude of the ripple yb can be fed back to control block 101 to update the control voltage u0. Specifically, the amplitude of the ripple yb can provide the speed ω and flux Φ used by the speed controller and flux controller.

[0050] The method includes, in step 107, demodulating the current y measured at step 106. Demodulating the current y includes separating the interference current yb S1 from the fundamental current ya. Demodulating the current y also includes determining the amplitude of the ripple yb of the interference current yb S1. Therefore, the amplitudes of the fundamental current ya and the ripple yb can be fed back to control step 101 to update the control voltage u0. As described above, the control system 22 is a closed-loop control system 22.

[0051] Figure 4 Block 107 shows the demodulated signal in more detail.

[0052] like Figure 4 As shown, demodulating the current y includes, in block 1071, applying a first finite impulse response (FIR) filter to the current signal y. The response of the FIR filter has a finite duration and is independent of the filter's previous input. The duration of the response from the first impulse response filter can therefore be adapted to one or more frequencies of the high-frequency signal S0. Thus, the finite impulse response filter can appropriately remove the interfering current yb S1 from the current y. The fundamental current ya can be extracted from the current y.

[0053] The demodulation current also includes, in step 1072, subtracting the fundamental current ya from the current y. Therefore, the interference current yb S1 is separated from the current y. The interference current yb S1 can be extracted from the current y.

[0054] The demodulated current y includes determining the amplitude of the ripple yb, which carries information about the state of the motor 14. Therefore, the actual speed ω and flux Φ of the speed controller and flux controller at step 101 can be determined based on the amplitude of the ripple yb.

[0055] Determining the amplitude of the ripple yb includes, in step 1073, calculating the zero-mean primitive S1 of the high-frequency signal S0. The zero-mean primitive S1 should be understood as a primitive that is zero in average over each period of the high-frequency signal S0.

[0056] Determining the amplitude of the ripple yb involves, in step 1074, multiplying the zero-mean primitive S1 of the high-frequency signal S0 by the interference current yb S1. It should be noted that multiplying the zero-mean primitive S1 of the high-frequency signal S0 by the interference current yb S1 also includes transposing the zero-mean primitive S1 of the high-frequency signal S0. Therefore, the ripple yb can be extracted from the interference current yb S1.

[0057] Determining the amplitude of the ripple yb includes, in step 1075, applying a second finite impulse response to the ripple yb. In other words, step 1075 includes taking the moving average of the interference current yb S1 multiplied by the zero-mean primitive S1. This moving average corresponds to the amplitude of the ripple yb. As described above, the finite impulse response filter is specifically adapted to one or more varying frequencies of the interference current yb S1.

[0058] Figure 5 An example is shown of a high-frequency signal S0 that can be injected into the control voltage u0 in step 102 and an interference current yb S1 induced as a result of the injected high-frequency signal S0.

[0059] Here, the high-frequency signal S0 is a square wave signal. When the high-frequency signal S0 is a square wave signal, the zero-mean primitive S1 of the high-frequency signal S0 can be easily calculated when demodulating the interference current yb S1 in step 1073.

[0060] As shown in the figure, the interference current yb S1 is a triangular signal, and the basic element S1 of the high-frequency signal S0 is modulated in amplitude by the ripple yb. As mentioned above, the ripple yb can be extracted using the basic element S1 of the injected high-frequency signal S0.

[0061] Furthermore, in the example shown, the frequency of the interference current yb S1 is high during the first duration T1 and low during the second duration T2. ​​The durations of the responses of the first and second finite impulse response filters applied in steps 1071 and 1075 can be adapted to the first duration T1 and then to the second duration T2. ​​Therefore, the current demodulation is adapted to one or more varying frequencies of the current y induced by the high-frequency signal S0.

[0062] This disclosure is not limited to the few examples described above, but is susceptible to variations available to those skilled in the art. For example, further signal processing blocks, such as amplification, filtering, and logic steps, may be implemented in the control method.

Claims

1. A method for sensorless control of an electric motor (14) implemented in a variable speed drive (12), wherein the method comprises: Determine the control voltage u0 to be applied to the motor (14); A high-frequency signal S0 is injected into the control voltage u0 to obtain an excitation voltage u1, wherein one or more frequencies of the high-frequency signal S0 vary with time; The excitation voltage u1 is applied to the motor (14); Measure the current y induced in the motor by the excitation voltage (u1), wherein the current y includes the fundamental current y induced by the control voltage u0. a and the interference current y induced by the high-frequency signal S0 b S1; Demodulating the current signal y, wherein demodulating the current signal y includes: A first finite impulse response filter is applied to the current signal y to extract the fundamental current y. a The duration of the response from the first finite impulse response filter varies according to one or more frequencies of the high-frequency signal S0, and Subtract the fundamental current y from the current signal y. a To extract the interference current y b S1; Determine the interference current y b S1 ripple y b The amplitude of the ripple y b The amplitude depends on the state of the motor.

2. The method according to claim 1, wherein, Determine the ripple y b The amplitude includes calculating the zero-mean primitive S1 of the high-frequency signal S0 and the interference current y b S1 multiplied by the zero-mean primitive S1.

3. The method according to claim 2, wherein, Determine the ripple y b The amplitude also includes the application of a second finite impulse response filter, wherein the duration of the response from the second finite impulse response filter varies according to one or more frequencies of the high-frequency signal S0.

4. The method according to any one of claims 1 to 3, wherein, One or more frequencies of the high-frequency signal S0 change randomly over time.

5. The method according to any one of claims 1 to 3, wherein, The high-frequency signal S0 is a sinusoidal signal.

6. The method according to any one of claims 1 to 3, wherein, The high-frequency signal S0 is a square wave signal.

7. The method according to any one of claims 1 to 3, wherein determining the control voltage u0 includes using a current controller, and applying the excitation voltage u1 to the motor includes using a pulse width modulation carrier u car The excitation voltage u1 is converted into a pulse width modulation voltage u PWM Furthermore, the fundamental frequency of the high-frequency signal S0 is determined by the frequency bandwidth of the current controller and the pulse width modulation carrier u. car Within the frequency range defined.

8. The method according to claim 7, wherein, The fundamental frequency of the high-frequency signal S0 varies between 250 Hz and 1 kHz.

9. The method according to any one of claims 1 to 3, wherein, The control voltage u0 is determined in the estimated rotor reference frame of the motor, and the high-frequency signal S0 is injected into the estimated rotor reference frame.

10. The method according to any one of claims 1 to 3, further comprising using the fundamental current y a and the ripple y b The amplitude of the voltage is updated to form a feedback loop by updating the control voltage u0.

11. A variable speed drive (12) for an electric motor (14), comprising a processor and a memory, the processor being configured to operate according to any one of the preceding claims.

12. An electric drive assembly (10) comprising a variable speed drive (12) according to claim 11 and an electric motor (14) controlled by the variable speed drive (12).

13. A computer-readable storage medium comprising instructions that, when executed by a processor, cause the processor to perform the method according to any one of claims 1 to 10.

Images