Inverter dead-time compensation method in high-frequency injection method position sensorless control

Abstract

The invention discloses an inverter dead-time compensation method in high-frequency injection method position sensorless control, and relates to the field of inverter dead-time compensation. The objective of the invention is to solve the problem of poor precision of a method for obtaining the position of a rotor of a permanent magnet synchronous motor. The method comprises the steps of 1, converting three-phase current of the permanent magnet synchronous motor into d-axis current and q-axis current, injecting positive and negative alternating square wave signals into the d-axis current, and extracting current error signals related to position errors from the q-axis current; 2, performing harmonic suppression processing on the current error signal by adopting an inverter nonlinear suppression module to obtain a processed current error signal; and step 3, obtaining a rotor estimation position from the processed current error signal by adopting a position phase-locked loop. The method is used for suppressing or eliminating the sixth harmonic generated by the dead zone of the inverter, so that the estimated position of the rotor is accurately obtained.

Description

Technical Field

[0001] This invention relates to the field of inverter dead-zone compensation. Background Technology

[0002] High-precision vector control of permanent magnet synchronous motors (PMSMs) requires real-time acquisition of rotor position information. The common practice is to install position sensors on the motor shaft system. However, these sensors increase the weight, size, and maintenance cost of the PMSM control system and are susceptible to environmental factors such as humidity and cold. Therefore, sensorless control strategies are currently employed, using electrical signals such as motor voltage and current to estimate the motor's position. Among these, the high-frequency square wave injection method is a widely used sensorless control strategy for PMSMs in the zero-speed and low-speed stages. The basic principle can be summarized as follows: injecting alternating positive and negative square wave signals into the estimated d-axis, extracting the high-frequency component related to the position error from the q-axis current, and sending it to the position observer to adjust the error to zero, thereby achieving rotor position estimation. However, to prevent the upper and lower switches of the same bridge arm in the inverter from conducting simultaneously, a dead zone is usually artificially added between the conduction signals of the upper and lower bridge arm switches. The addition of dead time will cause the high-frequency components generated in the square wave signal injection to be superimposed with the 6kth harmonic. Even if the position observer is used for adjustment, the error cannot be adjusted to 0. Therefore, the addition of the inverter dead time reduces the accuracy of rotor position estimation.

[0003] The impact of dead zones will be analyzed below.

[0004] Taking the A-phase bridge arm of the inverter as an example, the schematic diagram is as follows: Figure 1 As shown, the following are the divisions. <0 and We will discuss the two cases where the value is >0.

[0005] (1) >0

[0006] When switch S4 is turned off, due to the dead time, S1 has no turn-on signal and remains in the off state. At this time, the current path is D4-RL, and the output voltage is... ,in Given the diode's forward voltage drop, when switch S1 is turned on, the current path is S1-RL, and the output voltage is... ,in This is the on-state voltage drop of the switching transistor.

[0007] When S1 is off, S4 remains off. At this time, current flows through D4 via the path D4-RL, and the output voltage is... When S4 receives the conduction signal, the current path remains D4-RL, and the output voltage is... . Figure 2(a) shows the switching signal and output voltage waveforms under ideal conditions when >0 and when considering the influence of dead zone.

[0008] (2) <0

[0009] When switch S1 is turned off, S4 has no conduction signal and remains in the off state. The current path is LR-D1, and the output voltage is... When S4 is turned on, the current path is LR-S4, and the output voltage... .

[0010] After switch S4 is turned off, the current freewheels through D1, following the path LR-D1, and the output voltage is... .

[0011] Figure 2(b) shows the switching signal and output voltage waveforms under ideal conditions when <0 and when considering the dead zone effect.

[0012] As shown in Figure 2, the dead zone causes a difference between the actual voltage and the reference voltage, which in turn distorts the phase current.

[0013] Fourier decomposition of the output deviation in Figure 2 yields:

[0014] (1)

[0015] Because the motor uses a three-phase star connection, the output voltage will not contain the 3k harmonic. Therefore, due to the inverter dead zone, the phase current will eventually exhibit fifth and seventh harmonics, etc. Secondary harmonics.

[0016] Therefore, the current in phase A can be expressed as:

[0017] (2)

[0018] Expanding the above equation, we get:

[0019] (3)

[0020] in , .

[0021] Applying the Clarke and Park transformations, the dq-axis currents can be obtained as follows:

[0022] (4)

[0023] The above formula can be simplified to:

[0024] (5)

[0025] in: , , , .

[0026] Similarly, the high-frequency components generated during square wave signal injection will also be superimposed with 6k harmonics:

[0027] (6)

[0028] Based on coordinate transformation, the above equation is transformed to extract the current. Shaft system:

[0029] (7)

[0030] Based on the previous signal processing methods, the error signal can be obtained as follows:

[0031] (8)

[0032] in

[0033] Will When the PI controller is fed in and adjusted to 0, it can be seen that... Instead, there is a fixed angular deviation:

[0034] (9)

[0035] As can be seen from the above analysis, the inverter dead zone will cause the high-frequency response current to contain the 6kth harmonic, and ultimately cause the estimated angle to contain a deviation, which limits the operating performance of the sensorless control system. Summary of the Invention

[0036] The purpose of this invention is to solve the problem of poor accuracy in methods for obtaining the rotor position of a permanent magnet synchronous motor, and to propose a dead-zone compensation method for inverters in sensorless control using high-frequency injection.

[0037] A dead-zone compensation method for inverters in sensorless control using high-frequency injection, the method comprising the following:

[0038] Step 1: Convert the three-phase current of the permanent magnet synchronous motor into d-axis and q-axis currents. Inject alternating positive and negative square wave signals into the d-axis current and extract the current error signal related to the position error from the q-axis current. ;

[0039] Step 2: Use the inverter nonlinearity suppression module to suppress the current error signal. Harmonic suppression processing is performed to obtain the processed current error signal;

[0040] Step 3: Use a position phase-locked loop to obtain the rotor estimated position from the processed current error signal.

[0041] Preferably, in step 2, the specific process of obtaining the processed current error signal is as follows:

[0042] Current error signal Multiplying by the gain coefficient k, the output value is taken as the first output result. This result is multiplied by the differential constant, and the result is multiplied by the differential constant again. Then, it is multiplied by the center frequency of the second-order generalized integral. The result is added to the first output result, and the value is multiplied by the center frequency of the second-order generalized integral. The result is then multiplied by the differential constant and then compared with the current error signal. Perform a summation operation and output the value as the processed current signal.

[0043] Preferably, the center frequency of the second-order generalized integral is , ω is the angular frequency.

[0044] Preferably, in step 3, the specific process for obtaining the estimated rotor position is as follows:

[0045] After the processed current error signal is regulated by a PI controller, the estimated electrical angular velocity of the rotor is output. This electrical angular velocity is multiplied by a differential constant to obtain the estimated rotor position.

[0046] Preferably, the differential constant is .

[0047] The beneficial effects of this invention are:

[0048] This invention employs an inverter nonlinearity suppression module to suppress current error signals. Harmonic suppression processing is performed to suppress or eliminate the sixth harmonic caused by the dead zone, thereby improving the accuracy of rotor position estimation in the high-frequency injection method of permanent magnet synchronous motor. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the A-phase bridge arm of the inverter;

[0050] Figure 2(a) is Waveforms of switching signals and output voltage when >0 in ideal condition and when considering the effect of dead zone;

[0051] Figure 2(b) is Waveforms of switching signals and output voltage under ideal conditions and when considering the dead zone effect at <0;

[0052] Figure 3This is a schematic diagram of the generalized second-order integrator (SOGI).

[0053] Figure 4 The amplitude-frequency response and phase-frequency response curves of SOGI are shown.

[0054] Figure 5 This is a schematic diagram illustrating the principle of an inverter dead-zone compensation method in sensorless control using high-frequency injection. Detailed Implementation

[0055] 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.

[0056] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0057] Example:

[0058] A dead-zone compensation method for inverters in sensorless control using high-frequency injection, the method comprising the following:

[0059] Step 1: Convert the three-phase current of the permanent magnet synchronous motor into d-axis and q-axis currents. Inject alternating positive and negative square wave signals into the d-axis current and extract the current error signal related to the position error from the q-axis current. ;

[0060] Step 2: Use the inverter nonlinearity suppression module to suppress the current error signal. Harmonic suppression processing is performed to obtain the processed current error signal;

[0061] Step 3: Use a position phase-locked loop to obtain the rotor estimated position from the processed current error signal.

[0062] Specifically, the current error signal is the high-frequency component.

[0063] The inverter dead-time factor will generate a sixth harmonic error in the extracted high-frequency response current. To suppress the influence of the dead-time, the sixth harmonic needs to be suppressed or eliminated. Therefore, this embodiment uses a dead-time compensation method based on a generalized second-order integrator (SOGI) to improve the rotor position estimation accuracy. The structure of the generalized second-order integrator is shown in Figure 3, where the transfer function between the input and different outputs is:

[0064]

[0065] As can be seen from the transfer function, the port The notch characteristics of the input were achieved. By reasonably setting the center frequency and parameter k of SOGI, the position detection error information was obtained. The sixth harmonic component in the inverter can be eliminated, thereby suppressing the impact of the inverter dead zone on the accuracy of rotor position estimation.

[0066] With a fixed center frequency, SOGI has only one adjustable parameter k, making SOGI parameter tuning relatively simple. The amplitude-frequency response curves and phase-frequency response curves for different k values ​​are shown below, with the center frequency fixed. Figure 4 As shown, the arrow points in the direction of increasing k value.

[0067] Depend on Figure 4 It can be seen that the amplitude is significantly attenuated at the center frequency, while the amplitude is almost unaffected at other frequencies, thus meeting the requirements for notch filtering. Furthermore, as the value of k decreases, the notch filtering effect becomes better, but the corresponding bandwidth becomes narrower. In practical applications, the value of k should be chosen appropriately.

[0068] Further specifying, in step 2, the specific process of obtaining the processed current error signal is as follows:

[0069] Current error signal Multiplying by the gain coefficient k, the output value is taken as the first output result. This result is multiplied by the differential constant, and the result is multiplied by the differential constant again. Then, it is multiplied by the center frequency of the second-order generalized integral. The result is added to the first output result, and the value is multiplied by the center frequency of the second-order generalized integral. The result is then multiplied by the differential constant and then compared with the current error signal. Perform a summation operation and output the value as the processed current signal.

[0070] Further specifying, the center frequency of the second-order generalized integral is , ω is the angular frequency.

[0071] Further specifying, in step 3, the specific process for obtaining the estimated rotor position is as follows:

[0072] After the processed current error signal is regulated by a PI controller, the estimated electrical angular velocity of the rotor is output. This electrical angular velocity is multiplied by a differential constant to obtain the estimated rotor position.

[0073] Further specifying, the differential constant is .

[0074] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A method for dead-time compensation in sensorless control of high frequency injection, characterized in that, The method includes the following: Step 1: Convert the three-phase current of the permanent magnet synchronous motor into d-axis and q-axis currents. Inject alternating positive and negative square wave signals into the d-axis current and extract the current error signal related to the position error from the q-axis current. ; Step 2, using the inverter nonlinear suppression module to suppress the current error signal Harmonic suppression processing is performed to obtain a processed current error signal; Step 3: Use a position phase-locked loop to obtain the rotor estimated position from the processed current error signal.

2. The dead-time compensation method of an inverter in a high-frequency injection position sensorless control according to claim 1, characterized by, In step 2, the specific process of obtaining the processed current error signal is as follows: Current error signal Multiplying by the gain coefficient k, the output value is taken as the first output result. This result is multiplied by the differential constant, and the result is multiplied by the differential constant again. Then, it is multiplied by the center frequency of the second-order generalized integral. The result is added to the first output result, and the value is multiplied by the center frequency of the second-order generalized integral. The result is then multiplied by the differential constant and then compared with the current error signal. Perform a summation operation and output the value as the processed current signal.

3. The inverter dead-zone compensation method in sensorless control using high-frequency injection as described in claim 1, characterized in that, The second order generalized integral center frequency is , is the angular frequency.

4. The dead-time compensation method of an inverter in a high-frequency injection position sensorless control according to claim 1, characterized by, In step 3, the specific process for obtaining the estimated rotor position is as follows: After the processed current error signal is regulated by a PI controller, the estimated electrical angular velocity of the rotor is output. This electrical angular velocity is multiplied by a differential constant to obtain the estimated rotor position.

5. The dead-time compensation method of an inverter in a high-frequency injection position-sensorless control according to claim 1, characterized by, The differential constant is .

Images