Sensorless control method for PMSM based on stator winding neutral point voltage

Abstract

This invention relates to a sensorless control method for a PMSM (PMSM) based on the stator winding neutral point voltage. It uses three symmetrical resistors connected in parallel to reconstruct a virtual neutral point for the PMSM. The third harmonic is obtained by measuring the voltage difference between the stator winding neutral point and the virtual neutral point, and rotor position information is extracted using SOGI-FLL-WPF technology. This invention has low manufacturing cost, does not require the injection of additional high-frequency signals or dependence on motor parameters, and only requires the motor to have a star-connected stator winding.

Description

Technical Field

[0001] This invention relates to a sensorless control method for PMSMs based on the stator winding neutral point voltage, belonging to the field of sensorless control of permanent magnet synchronous motors (PMSMs), and particularly to a sensorless control method based on the voltage difference between the stator winding neutral point and the virtual neutral point and a second-order generalized integrator-based frequency-locked loop with prefilter (SOGI-FLL-WPF) technology. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) are widely used in aerospace, electric vehicles, and home appliances due to their advantages such as simple structure, high power density, and good reliability. The classic control method for PMSMs is field-oriented control (FOC). However, this method requires precise rotor position and speed information, and usually requires the installation of position sensors, such as Hall sensors and quadrature encoders. This increases the cost, size, and weight of the system, and reduces its reliability.

[0003] To reduce the cost and improve the reliability of permanent magnet synchronous motor drive systems, sensorless control methods have been widely adopted. These methods can be broadly categorized into two types: rotor salient polarity-based and model-based control methods. However, rotor salient polarity-based control methods require the injection of high-frequency signals, which generates electromagnetic noise. Furthermore, high-frequency injection methods are difficult to apply to medium- and high-speed applications because the bus voltage limits the amplitude of the injected voltage. Model-based methods offer high position estimation accuracy at medium and high speeds, but they are susceptible to internal and external disturbances. Internal disturbances are mainly variations in motor parameters, while external disturbances are primarily load, inverter nonlinearity, sampling bias, and harmonics. As motor speed decreases, these disturbances severely degrade position estimation accuracy. Summary of the Invention

[0004] To address the aforementioned problem of sensorless control of permanent magnet synchronous motors, this invention proposes a sensorless control method based on the neutral point voltage of the stator winding. This method does not require the injection of additional high-frequency signals or dependence on motor parameters. It directly detects the voltage difference between the neutral point and the virtual neutral point, and then uses a second-order generalized integrator-based frequency-locked loop with prefilter (SOGI-FLL-WPF) technique to extract rotor position information, thereby achieving sensorless control.

[0005] The PMSM sensorless control method based on the stator winding neutral point voltage includes the following steps:

[0006] Step 1: Rotor pre-positioning: Setting the current By outputting a fixed voltage vector through the current loop PI controller, the rotor d-axis is aligned with the permanent magnet synchronous motor winding a, thus avoiding the problems of open-loop start failure and reverse rotation caused by If control.

[0007] Step 2: If open-loop control, with current closed-loop control, employs a uniform acceleration start-up strategy to accelerate the motor speed to the target set speed; the relationship between the virtual synchronous coordinate system and the actual rotor dq-axis coordinate system in the If open-loop control is: the virtual coordinate system leads the actual coordinate system by π / 2.

[0008] Step 3: After the If open-loop control stabilizes, the third harmonic back electromotive force e3 is extracted by directly detecting the voltage difference between the neutral point and the virtual neutral point of the stator winding of the permanent magnet synchronous motor; by introducing a virtual neutral point and then directly detecting the voltage difference between the neutral point and the virtual neutral point, the third harmonic back electromotive force is extracted.

[0009] Obtaining the Third Harmonic Back EMF. Since the magnetic flux density of the rotor permanent magnet in the air gap around the motor is a non-ideal sinusoidal wave, the air gap flux contains integer multiples of the third harmonic, with the third harmonic accounting for the largest proportion of the total harmonics. This third harmonic flux will generate a third harmonic back EMF at the stator winding neutral point. Therefore, we use three symmetrical resistors connected in parallel on the motor phase lines to reconstruct a virtual neutral point of the PMSM. By directly detecting the voltage difference between the permanent magnet synchronous motor neutral point and the virtual neutral point, the third harmonic back EMF can be effectively extracted.

[0010] Step 4: The third harmonic back EMF signal is filtered through a pre-filter, which is a frequency adaptive bandpass filter. This allows the SOGI-FLL-WPF technology to achieve a better trade-off between speed and accuracy compared to the standard SOGI-FLL. It is also an SOGI-based quadrature signal generator, but only uses the output of the α axis.

[0011] Step 5: Construct the signal and its orthogonal signal obtained from filtering in Step 4 using the SOGI module, and simultaneously use FLL technology to track the electrical angular frequency of the signal;

[0012] The filtered third harmonic back electromotive force e′3 is obtained, but the rotor position cannot be directly calculated from a single signal; therefore, an orthogonal signal needs to be constructed. Furthermore, the electrical angular frequency needs to be tracked in real time to match the speed change. Therefore, the filtered signal is passed through the SOGI module to obtain the estimated position. Signals and their orthogonal signals Simultaneously, a frequency-locked loop (FLL) is used to track and estimate the signal, yielding three times the electrical angular frequency. When the SOGI-FLL module is in steady state, the estimated electrical angular frequency Equal to the actual ω 3e ,so and The transfer function between e′3 and e′3 can be expressed as:

[0013]

[0014] Where k2 is the gain coefficient of the SOGI-FLL module.

[0015] Step 6: Normalize the two orthogonal signals obtained in Step 5, and obtain the estimated electrical angle value through the arctangent function, then perform frequency conversion;

[0016] Then and After unitization, we obtain and Finally, the estimated electrical angle was calculated using the arctangent function. Since the rotor position information is extracted from the third harmonic back electromotive force, this angle is three times the actual electrical angular frequency. Converted to actual electrical angle θ e ;

[0017] Step 7: Replace the electrical angle value from Step 6 with the angle generated by the If open-loop control, calculate the actual speed of the motor through differentiation, and introduce a speed loop to achieve sensorless control.

[0018] The rotor position θ obtained in step sixe Replace the angle generated in the If open-loop control in step two. via θ e Calculated velocity And A speed loop is introduced for closed-loop speed control.

[0019] Step two specifically includes:

[0020] A uniform acceleration start-up strategy is adopted, and the magnitude of the d-axis current is set to remain at i. d =0, the q-axis current is a constant starting current, passing through a given current angular velocity ω e The position angle controlled by the motor is obtained by integration. Given a position, in the initial stage, a coordinate axis d′q′ that leads the actual coordinate system by π / 2 is given as a virtual synchronous coordinate system. The PMSM torque equation and motion equation are rewritten as follows:

[0021]

[0022] In equations (1) and (2): T e For electromagnetic torque, n p Let ψ be the extreme logarithm. f For permanent magnet flux linkage, θ γ The angular error between the coordinate systems is J, where J is the moment of inertia and ω is ω. r T is the mechanical angular velocity. l Where is the load torque, and B is the coefficient of friction.

[0023] Step four specifically includes:

[0024] After passing through a frequency adaptive bandpass filter, let... It is a constant, and the transfer function between e′3 extracted by the pre-filter and the input signal e3 can be expressed as:

[0025]

[0026] Where k1 is the gain coefficient of the pre-filter. The electrical angular frequency of the signal. The optimal choice corresponds to the damping coefficient. Ensure that the pre-filter achieves the optimal balance between overshoot and settling time when extracting the signal.

[0027] Step five specifically includes:

[0028] The parameters in SOGI are selected by linearizing the SOGI model:

[0029]

[0030] The poles of the closed-loop transfer function are all real numbers, corresponding to the case where the damping coefficient is equal to or greater than 1. When the damping coefficient is greater than 1, the response speed is too slow. Choosing a damping coefficient equal to the unit damping factor corresponds to the transfer function having the same poles, so k1 = k2. The characteristic polynomial of equation (8) is:

[0031]

[0032] Since equation (10) is a third-order polynomial with at least one real pole, it can also be expressed as:

[0033]

[0034] Where ζ and ω n Let represent the damping factor and natural frequency, respectively. γ is the factor that determines the location of the real poles. According to equation (11), the coefficients of s of the same order are equal on both sides, and let The gain λ of the FLL can then be expressed as:

[0035]

[0036] From equations (11) and (12), we can see that γ can be expressed as:

[0037]

[0038] Step six specifically includes:

[0039] The two orthogonal signals generated by the SOGI module are normalized, and the rotor electrical angle is obtained through the arctangent function. When rotating counterclockwise, 2π / 3 and 4π / 3 are added to the second and third cycles of three times the electrical angle frequency, respectively; when rotating clockwise, 4π / 3 and 2π / 3 are added to the first and second cycles, respectively. The mathematical expression for this frequency conversion is shown in formula (14):

[0040]

[0041] Among them, D ir As the direction symbol, D becomes D when rotated counterclockwise. ir The value is 0; when rotated clockwise, D... ir =1, ω f This is the period of electrical angle.

[0042] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0043] This invention uses three symmetrical resistors connected in parallel to reconstruct the virtual neutral point of the PMSM (Motor Motor Module). The third harmonic is obtained by measuring the voltage difference between the stator winding neutral point and the virtual neutral point. Rotor position information is extracted using SOGI-FLL-WPF technology. First, the third harmonic signal is pre-filtered. Since the rotor position cannot be directly calculated from a single signal, an orthogonal signal is constructed using the SOGI module. Furthermore, the FLL module tracks the electrical angular frequency in real time to match speed changes. Then, the two orthogonal signals are normalized and the rotor position information is obtained using the arctangent function. Because the angle is obtained from the third harmonic, the obtained electrical angular frequency is three times the actual electrical angular frequency. This electrical angle is then converted to the actual electrical angle. Finally, this angle replaces the given angle in the If control, and the motor speed is calculated and incorporated into the speed loop, thus achieving sensorless control. This invention has low manufacturing cost, does not require additional high-frequency signals or dependence on motor parameters, only requires a star-connected stator winding, and the sensor part only uses basic circuitry without the need for position sensors such as Hall sensors and quadrature encoders, resulting in low manufacturing costs.

[0044] This invention proposes a novel sensorless control method based on the neutral point voltage of a permanent magnet synchronous motor and a second-order generalized integrator-based frequency-locked loop with prefilter (SOGI-FLL-WPF). Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a rotor coordinate system diagram of the PMSM sensorless control method based on the neutral point voltage of the stator winding according to the present invention.

[0047] Figure 2 This is a block diagram of the SOGI-FLL-WPF method for sensorless control of PMSM based on the neutral point voltage of the stator winding according to the present invention.

[0048] Figure 3 This is a block diagram of the sensorless control structure of the PMSM sensorless control method based on the neutral point voltage of the stator winding according to the present invention.

[0049] Figure 4 This is a schematic diagram of the three-fold electrical angular frequency conversion of the PMSM sensorless control method based on the neutral point voltage of the stator winding according to the present invention.

[0050] Figure 5 This is a simulation result of sensorless control loading for the PMSM sensorless control method based on the neutral point voltage of the stator winding according to the present invention.

[0051] Figure 6 This is a simulation result of sensorless speed-up control based on the PMSM sensorless control method of the stator winding neutral point voltage according to the present invention.

[0052] Figure 5 (a) and Figure 6 (a) is the rotational speed waveform. Figure 5 (b) and Figure 6 (b) shows the electromagnetic torque waveform. Figure 5 (c) and Figure 6 (c) is the waveform of the third harmonic back electromotive force. Figure 5 (d) and Figure 6 (d) is the electrical angle waveform. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] Reference Figure 3 The diagram shown illustrates the structural block diagram of the present invention. It includes five modules: input, control, drive, permanent magnet synchronous motor, and DC power supply. The input module includes a mechanical reference speed. and open-loop reference current The control module includes open-loop If control and closed-loop FOC control; the drive module includes PWM output, inverter, amplifier, and AD converter.

[0055] The PMSM sensorless control method based on the stator winding neutral point voltage includes the following steps:

[0056] Step 1: Rotor pre-positioning: Setting the current By outputting a fixed voltage vector through the current loop PI controller, the rotor d-axis is aligned with the permanent magnet synchronous motor winding a, thus avoiding the problems of open-loop start failure and reverse rotation caused by If control.

[0057] Step 2: For open-loop control, set the target open-loop speed to 600 r / min and the acceleration to 4000 r / min. Convert the speed to electrical angular velocity and integrate it to obtain the motor's controlled position angle. Then, add the obtained electrical angle to establish a coordinate axis leading the actual coordinate system. Set the current closed-loop control and speed open-loop control to forcefully pull the motor to the set speed.

[0058] Step 3: After the If open-loop control stabilizes, the third harmonic back electromotive force e3 is extracted by directly detecting the voltage difference between the neutral point and the virtual neutral point of the stator winding of the permanent magnet synchronous motor; by introducing a virtual neutral point and then directly detecting the voltage difference between the neutral point and the virtual neutral point, the third harmonic back electromotive force is extracted.

[0059] Obtaining the Third Harmonic Back EMF. Since the magnetic flux density of the rotor permanent magnet in the air gap around the motor is a non-ideal sinusoidal wave, the air gap flux contains integer multiples of the third harmonic, with the third harmonic accounting for the largest proportion of the total harmonics. This third harmonic flux will generate a third harmonic back EMF at the stator winding neutral point. Therefore, we use three symmetrical resistors connected in parallel on the motor phase lines to reconstruct a virtual neutral point of the PMSM. By directly detecting the voltage difference between the permanent magnet synchronous motor neutral point and the virtual neutral point, the third harmonic back EMF can be effectively extracted.

[0060] Step 4: The third harmonic back EMF signal is filtered through a pre-filter, which is a frequency adaptive bandpass filter. This allows the SOGI-FLL-WPF technology to achieve a better trade-off between speed and accuracy compared to the standard SOGI-FLL. It is also an SOGI-based quadrature signal generator, but only uses the output of the α axis.

[0061] Step 5: Construct the signal and its orthogonal signal obtained from filtering in Step 4 using the SOGI module, and simultaneously use FLL technology to track the electrical angular frequency of the signal;

[0062] The filtered third harmonic back electromotive force e′3 is obtained, but the rotor position cannot be directly calculated from a single signal; therefore, an orthogonal signal needs to be constructed. Furthermore, the electrical angular frequency needs to be tracked in real time to match the speed change. Therefore, the filtered signal is passed through the SOGI module to obtain the estimated position. Signals and their orthogonal signals Simultaneously, a frequency-locked loop (FLL) is used to track and estimate the signal, yielding three times the electrical angular frequency. When the SOGI-FLL module is in steady state, the estimated electrical angular frequency Equal to the actual ω 3e ,so and The transfer function between e′3 and e′3 can be expressed as:

[0063]

[0064] Where k2 is the gain coefficient of the SOGI-FLL module.

[0065] Step 6: Normalize the two orthogonal signals obtained in Step 5, and obtain the estimated electrical angle value through the arctangent function, then perform frequency conversion;

[0066] Then and After unitization, we obtain and Finally, the estimated electrical angle was calculated using the arctangent function. Since the rotor position information is extracted from the third harmonic back electromotive force, this angle is three times the actual electrical angular frequency. Converted to actual electrical angle θ e ;

[0067] Step 7: Replace the electrical angle value from Step 6 with the angle generated by the If open-loop control, calculate the actual speed of the motor through differentiation, and introduce a speed loop to achieve sensorless control.

[0068] The rotor position θ obtained in step six e Replace the angle generated in the If open-loop control in step two. via θ e Calculated velocity And A speed loop is introduced for closed-loop speed control.

[0069] Step two specifically includes:

[0070] A uniform acceleration start-up strategy is adopted, and the magnitude of the d-axis current is set to remain at i. d =0, the q-axis current is a constant starting current, passing through a given current angular velocity ω e The position angle controlled by the motor is obtained by integration. Given a position, in the initial stage, a coordinate axis d′q′ that leads the actual coordinate system by π / 2 is given as a virtual synchronous coordinate system. The PMSM torque equation and motion equation are rewritten as follows:

[0071]

[0072] In equations (1) and (2): T e For electromagnetic torque, n p Let ψ be the extreme logarithm. f For permanent magnet flux linkage, θ γ The angular error between the coordinate systems is J, where J is the moment of inertia and ω is ω. rT is the mechanical angular velocity. l Where is the load torque, and B is the coefficient of friction.

[0073] Step four specifically includes:

[0074] After passing through a frequency adaptive bandpass filter, let... It is a constant, and the transfer function between e′3 extracted by the pre-filter and the input signal e3 can be expressed as:

[0075]

[0076] Where k1 is the gain coefficient of the pre-filter. The electrical angular frequency of the signal. The optimal choice corresponds to the damping coefficient. Ensure that the pre-filter achieves the optimal balance between overshoot and settling time when extracting the signal.

[0077] Step five specifically includes:

[0078] get and The transfer function cannot reflect the frequency and angle tracking characteristics of SOGI-FLL-WPF. (This is achieved through...) Figure 2 Perform linearization, and It can be represented as:

[0079]

[0080] Since the poles of the closed-loop transfer function (8) are all real numbers, this corresponds to the case where the damping coefficient is equal to or greater than 1. However, when the damping coefficient is greater than 1, the dynamic response of the transfer function (8) is overdamped, and the response speed is too slow. Therefore, the damping coefficient is chosen to be equal to the unit damping factor, which corresponds to the transfer function (8) having the same poles, so k1 = k2. The characteristic polynomials of equations (8) and (9) are:

[0081]

[0082] Since equation (10) is a third-order polynomial with at least one real pole, it can also be expressed as:

[0083]

[0084] Where ζ and ω n Let represent the damping factor and natural frequency, respectively. γ is the factor that determines the location of the real poles. According to equation (11), the coefficients of s of the same order are equal on both sides, and let The gain λ of the FLL can then be expressed as:

[0085]

[0086] From equations (11) and (12), we can see that γ can be expressed as:

[0087]

[0088] Step six specifically includes:

[0089] The two orthogonal signals generated by the SOGI module are normalized, and then the rotor electrical angle is obtained through the arctangent function. Figure 4 To illustrate the conversion of this angle to an actual electrical angle, the diagram includes both counter-clockwise and clockwise rotation of the motor. When rotating counter-clockwise, 2π / 3 and 4π / 3 are added to the second and third cycles, respectively, at three times the electrical angular frequency; when rotating clockwise, 4π / 3 and 2π / 3 are added to the first and second cycles, respectively. The mathematical expression for this frequency conversion is shown in the formula:

[0090]

[0091] Among them, D ir As the direction symbol, D becomes D when rotated counterclockwise. ir The value is 0; when rotated clockwise, D... ir =1, ω f This is the period of electrical angle.

[0092] Step seven specifically includes:

[0093] Closed-loop FOC control. When the motor speed reaches the reference speed set in the open loop and remains stable, the switch S... w1 and S w2 The state is set to 1 to switch from open-loop control to closed-loop control, that is, using the estimated electrical angle. replace and mechanical speed A velocity loop is introduced as a velocity feedback mechanism.

[0094] To verify the control performance of the PMSM sensorless control method based on the stator winding neutral point voltage under sudden load increase and speed change conditions, simulation tests were conducted under the following conditions: the load was increased to 0.1 at 1.4s, and the motor speed was uniformly increased from 1500 r / min to 1700 r / min at 1.8s. The simulation results are as follows: Figure 5 and Figure 6 As shown, this includes speed tracking, electromagnetic torque variation, third harmonic back EMF tracking, and rotor position error waveform.

[0095] Table 1 Parameters of Permanent Magnet Synchronous Motor

[0096] parameter numerical values parameter numerical values Extreme logarithm 2 bus voltage 24V Magnetic Link 0.0124Wb Rated current 5.5A d-axis inductance 2.5e-4H Rated torque 0.2 N·m q-axis inductor 3.5e-4H Rated speed 3900r / min

[0097] Figure 5 The simulated waveform of the motor at 1500 r / min speed, increasing from no-load to a load of 0.1, is displayed. Figure 5 (a) shows that when the load increases to 0.1, the actual motor speed briefly drops to 1542 r / min and recovers to the reference speed after 0.135 seconds, indicating that the method has good anti-interference ability. Figure 5 (c) It can be seen that the output signal of the SOGI-FLL-WPF module can quickly track the third harmonic back EMF signal, indicating that the technology has a good dynamic tracking effect on the third harmonic back EMF signal, so that the error between the estimated rotor position and the actual rotor position when the motor is in steady state is almost 0.

[0098] Figure 6 The simulated waveforms are shown when the reference speed is increased to 1700 r / min. (By...) Figure 6 (a) It can be seen that the actual speed of the motor can quickly track the reference speed, with an adjustment time of 0.2s. Figure 6 (c) It can be seen that the output signal of the SOGI-FLL-WPF module can also dynamically track the third harmonic back electromotive force signal very well.

[0099] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and these variations still fall within the protection scope of the present invention.

Claims

1. A sensorless control method for PMSM based on stator winding neutral point voltage, characterized in that: Includes the following steps: Step 1: Rotor pre-positioning: Setting the current A fixed voltage vector is output by the current loop PI controller to align the rotor d-axis with the permanent magnet synchronous motor winding a. Step 2: If open-loop control, control is current closed-loop control, using a uniform acceleration start strategy to accelerate the motor speed to the target set speed; Step 3: After the If open-loop control stabilizes, the third harmonic back electromotive force e3 is extracted by directly detecting the voltage difference between the neutral point and the virtual neutral point of the stator winding of the permanent magnet synchronous motor. Step 4: Filter the third harmonic back EMF signal through a pre-filter; Step 5: Construct the signal and its orthogonal signal obtained from filtering in Step 4 using the SOGI module, and simultaneously use FLL technology to track the electrical angular frequency of the signal; Step five specifically includes: The parameters in SOGI are selected by linearizing the SOGI model: The poles of the closed-loop transfer function are all real numbers, corresponding to the case where the damping coefficient is equal to or greater than 1. When the damping coefficient is greater than 1, the response speed is too slow. Choosing a damping coefficient equal to the unit damping factor corresponds to the transfer function having the same poles, so k1 = k2. The characteristic polynomial of equation (8) is: Since equation (9) is a third-order polynomial with at least one real pole, it can also be expressed as: Where ζ and ω n Let represent the damping factor and natural frequency, respectively. γ is the factor that determines the location of the real poles. According to equation (10), the coefficients of s of the same order are equal on both sides, and let The gain λ of the FLL can then be expressed as: From equations (10) and (11), we can see that γ can be expressed as: Step 6: Normalize the two orthogonal signals obtained in Step 5, and obtain the estimated electrical angle value through the arctangent function, then perform frequency conversion; Step 7: Replace the electrical angle value from Step 6 with the angle generated by the If open-loop control, calculate the actual speed of the motor through differentiation, and introduce a speed loop to achieve sensorless control.

2. The PMSM sensorless control method based on stator winding neutral point voltage according to claim 1, characterized in that: Step two specifically includes: A uniform acceleration start-up strategy is adopted, and the magnitude of the d-axis current is set to remain at i. d =0, the q-axis current is a constant starting current, passing through a given current angular velocity ω e The position angle controlled by the motor is obtained by integration. Given a position, in the initial stage, a coordinate axis d′q′ that leads the actual coordinate system by π / 2 is given as a virtual synchronous coordinate system. The PMSM torque equation and motion equation are rewritten as follows: In equations (1) and (2): T e For electromagnetic torque, n p Let ψ be the extreme logarithm. f For permanent magnet flux linkage, θ γ The angular error between the coordinate systems is J, where J is the moment of inertia and ω is ω. r T is the mechanical angular velocity. l Where is the load torque and B is the coefficient of friction.

3. The PMSM sensorless control method based on stator winding neutral point voltage according to claim 1, characterized in that: Step four specifically includes: After passing through a frequency adaptive bandpass filter, let... It is a constant, and the transfer function between e′3 extracted by the pre-filter and the input signal e3 can be expressed as: Where k1 is the gain coefficient of the pre-filter. The electrical angular frequency of the signal. The optimal choice corresponds to the damping coefficient. Ensure that the pre-filter achieves the optimal balance between overshoot and settling time when extracting the signal.

4. The PMSM sensorless control method based on stator winding neutral point voltage according to claim 1, characterized in that: Step six specifically includes: The two orthogonal signals generated by the SOGI module are normalized, and the rotor electrical angle is obtained through the arctangent function. When rotating counterclockwise, 2π / 3 and 4π / 3 are added to the second and third cycles of three times the electrical angle frequency, respectively; when rotating clockwise, 4π / 3 and 2π / 3 are added to the first and second cycles, respectively. The mathematical expression for this frequency conversion is shown in formula (14): Among them, D ir As the direction symbol, D becomes D when rotated counterclockwise. ir The value is 0; when rotated clockwise, D... ir =1, ω f This is the period of electrical angle.

Images