Nonlinear dead-time compensation method for online variable-coefficient silicon carbide inverter

Abstract

The invention provides an online variable coefficient silicon carbide inverter nonlinear dead zone compensation method, and belongs to the technical field of alternating current servo driving. The method comprises the following steps: modeling a built-in permanent magnet synchronous motor IPMSM; estimating the flux linkage of the built-in permanent magnet synchronous motor IPMSM to obtain the motor flux linkage observed by the online observer in real time; judging the polarity of the three-phase stator current to obtain a three-phase current phase angle; calculating to obtain an average error voltage compensation coefficient of the silicon carbide inverter; and based on the three-phase current phase angle and the average error voltage compensation coefficient of the silicon carbide inverter, calculating to obtain an improved three-phase online variable coefficient silicon carbide inverter dead-zone compensation voltage, and compensating the nonlinear dead-zone effect of the silicon carbide inverter by using the improved three-phase online variable coefficient silicon carbide inverter dead-zone compensation voltage. According to the invention, the problem of phase current harmonic distortion caused by the nonlinear dead-zone effect of the silicon carbide inverter is solved.

Description

Technical Field

[0001] This invention belongs to the field of AC servo drive technology, and particularly relates to an online variable coefficient silicon carbide inverter nonlinear dead zone compensation method. Background Technology

[0002] Silicon carbide (SiC) inverters and integrated permanent magnet synchronous motors (IPMSMs) offer advantages such as high efficiency and high power density, leading to their widespread application in electric vehicles. When using a SiC voltage source inverter to drive a motor, a certain dead time needs to be added to prevent short circuits in the inverter's upper and lower bridge power devices, which could damage them. The nonlinearity of the inverter caused by the artificially added dead time, the on-state voltage drop of the power devices, and the switching delay of the power devices leads to an increase in current harmonic content. This phenomenon is known as the dead-time effect, which increases motor output torque and speed pulsation, resulting in decreased motor performance. Furthermore, the dead-time effect intensifies with increasing switching frequency of the SiC inverter, leading to increased harmonic distortion of the motor phase current. The harmonic problems caused by the dead-time effect at high switching frequencies of SiC inverters are even more severe. Dead-time elimination methods compensate for the dead-time effect by adding additional circuit components, achieving high compensation accuracy but also increasing the system's hardware cost. Offline measurement and lookup table compensation methods reduce the impact of dead-time effects by offline measurement of the switching delay time and compensation voltage of power devices under different load currents. However, these methods have poor real-time performance and require extensive offline testing. One of the most widely used methods is the dead-time compensation method based on the average error voltage model, which has been widely applied in multilevel converters and parallel inverters. Because the switching delay and on-state voltage drop of silicon carbide inverter power devices are time-varying, it is difficult to establish an accurate inverter nonlinear model, resulting in limited compensation effectiveness for traditional dead-time compensation methods based on the average error voltage model. To improve the accuracy of the inverter compensation voltage model, some researchers have studied the influence of parasitic capacitance on the switching process of power devices to improve the dead-time compensation effect. However, the compensation model is complex, and the parasitic capacitance changes with the drain-source voltage of the device. Summary of the Invention

[0003] To address the aforementioned shortcomings of existing technologies, this invention provides a precise online variable-coefficient inverter nonlinearity dead-zone compensation method that is independent of silicon carbide inverter power device parameters. This method reduces current harmonic distortion caused by the dead-zone effect in silicon carbide inverters and improves motor output performance. It solves the following problems: While existing traditional dead-zone compensation methods based on average error voltage are simple, their compensation accuracy is limited. Some precise compensation methods employ offline measurement and lookup table methods, but these require extensive offline testing and have poor real-time performance. Dead-zone elimination methods require additional hardware circuitry, increasing system costs. While precise compensation methods based on the nonlinear model of silicon carbide inverters do not require additional hardware circuitry, they require accurate silicon carbide inverter power device parameters, such as parasitic capacitance. These parameters are too small and change in real-time with the drain-source voltage of the devices, making them difficult to measure.

[0004] To achieve the above objectives, the technical solution adopted by this invention is: an online variable coefficient silicon carbide inverter nonlinear dead zone compensation method, comprising the following steps: Modeling of the built-in permanent magnet synchronous motor (IPMSM); Based on the modeling results, the flux linkage of the built-in permanent magnet synchronous motor (IPMSM) is estimated, and the motor flux linkage observed in real time by the online observer is obtained. The polarity of the three-phase stator current is determined to obtain the phase angle of the three-phase current; Based on the motor flux observed in real time by the online observer, the average error voltage compensation coefficient of the silicon carbide inverter is calculated, where the average error voltage compensation coefficient of the silicon carbide inverter is a dynamic variable. Based on the three-phase current phase angle and the average error voltage compensation coefficient of the silicon carbide inverter, the improved dead-zone compensation voltage of the three-phase online variable coefficient silicon carbide inverter is calculated, and the nonlinear dead-zone effect of the silicon carbide inverter is compensated by the improved dead-zone compensation voltage of the three-phase online variable coefficient silicon carbide inverter.

[0005] The beneficial effects of this invention are: it avoids directly establishing a complex nonlinear error compensation model for silicon carbide inverters, treats the silicon carbide inverter as a black box, and uses an online observer to monitor the motor flux linkage in real time. To address the problem of misjudgment caused by repeated zero-crossing polarity jumps in the motor stator current, the relationship between the current vector angle and the rotor position angle is derived without adding additional hardware circuitry. Based on the current vector angle and the average error voltage compensation coefficient of the silicon carbide inverter calculated by the PI controller, online adaptive compensation is performed for the nonlinear dead-zone effect of the silicon carbide inverter, thus solving the phase current harmonic distortion problem caused by the nonlinear dead-zone effect of the silicon carbide inverter.

[0006] Furthermore, the expression for modeling the built-in permanent magnet synchronous motor (IPMSM) is as follows: ; ; ; p= ; in, u α and u β This represents the components of the stator winding phase voltage in a two-phase stationary coordinate system. i α and i β This represents the components of the stator winding phase current in a two-phase stationary coordinate system. L α and L β This represents the component of the stator winding phase inductance in a two-phase stationary coordinate system. L d and L q These represent the direct-axis inductance and quadrature-axis inductance of the stator winding, respectively. R This represents the phase resistance of the stator winding. oh e Indicates the rotor's electrical angular velocity. i e Indicates the rotor electrical angle, Indicates the magnetic flux linkage of the rotor permanent magnet. p Describes the differential operator. Indicates average inductance. Indicates half-differential inductance. Indicates coupled inductance. Represents the differential symbol. t This represents the differential time.

[0007] Furthermore, the expression for the motor flux linkage observed in real time by the online observer is: ; ; ; ; ; in, The derivative of the magnetic flux linkage observation. Represents the stator voltage vector. Represents the stator current vector. T This indicates the transpose operation. Indicates the flux linkage observer gain. Indicates the magnetic flux linkage observation value. Represents the cost function, The derivative of the observed flux linkage of the motor, This represents the observed value of the motor flux linkage. Represents the d-axis current. This represents the error driving function of the online observer.

[0008] Furthermore, the polarity of the three-phase stator current is determined to obtain the phase angle of the three-phase current, which is as follows: Define stator current vector i s The angle between the d-axis and the d-axis is i s ; According to the definition of the rotor dq synchronous rotating coordinate system, the d-axis and the rotor N pole position coincide. When the rotor position angle is... i e At that time, according to the inverse Park transform formula, we get: ; in, a Phase current i α Expressed by trigonometric function formulas: ; ; in, Represents the d-axis current. Indicates the rotor electrical angle, Represents the q-axis current. Represents the stator current vector. d a express a Phase current phase angle.

[0009] Based on the symmetrical distribution of the three-phase currents in the stator winding, the remaining two-phase currents are obtained. i b , i c phase angle d b and d c ,in, d a , d b and d c This represents the phase angle of the three-phase current.

[0010] Furthermore, phase angle d b and d c The expression is as follows: ; .

[0011] Furthermore, the average error voltage compensation coefficient of the silicon carbide inverter is calculated as follows: Based on the motor flux linkage observed in real time by the online observer, the estimated value of the rotor permanent magnet flux linkage is calculated, and the difference between the estimated value of the rotor permanent magnet flux linkage and the permanent magnet flux linkage constant is calculated. Based on the difference calculated, the average error voltage compensation coefficient of the silicon carbide inverter is obtained through the PI controller. k c .

[0012] Furthermore, the average error voltage compensation coefficient of the silicon carbide inverter k c The expression is as follows: ; in, This represents the proportionality coefficient. Represents the integral coefficient. Represents the Laplace operator, Indicates the magnetic flux linkage of the rotor permanent magnet. This represents the estimated value of the rotor permanent magnet flux linkage.

[0013] Furthermore, the expression for the dead-time compensation voltage of the improved three-phase online variable coefficient silicon carbide inverter is as follows: ; in, This indicates the dead-zone compensation voltage of the improved three-phase online variable coefficient silicon carbide inverter. This represents the average error voltage compensation coefficient for silicon carbide inverters. This indicates the switching frequency of the silicon carbide inverter. Indicates dead time. Indicates bus voltage. Represents a symbolic function. This indicates the phase angle of the three-phase current.

[0014] The beneficial effect of the above-mentioned further solution is that the improved dead-zone compensation voltage of the three-phase online variable coefficient silicon carbide inverter can be obtained through the above method, so as to adaptively compensate for the nonlinear dead-zone effect of the inverter. Attached Figure Description

[0015] Figure 1 This is a flowchart of the method of the present invention.

[0016] Figure 2 The average error voltage compensation coefficient for silicon carbide inverters k c Block diagram of the calculation principle.

[0017] Figure 3(a) shows the experimental waveform of the motor phase current without dead zone compensation.

[0018] Figure 3(b) is a Fourier analysis diagram of the motor phase current without dead zone compensation.

[0019] Figure 4(a) shows the experimental waveform of motor phase current based on average error voltage dead zone compensation.

[0020] Figure 4(b) is a Fourier analysis diagram of the motor phase current based on average error voltage dead zone compensation.

[0021] Figure 5(a) is an experimental waveform of the motor phase current using the dead zone compensation method proposed in this invention.

[0022] Figure 5(b) is a Fourier analysis diagram of the motor phase current using the dead zone compensation method proposed in this invention. Detailed Implementation

[0023] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0024] Example like Figure 1 As shown, this invention provides an online variable coefficient silicon carbide inverter nonlinear dead zone compensation method, the implementation method of which is as follows: S1. Model the built-in permanent magnet synchronous motor (IPMSM); In this embodiment, firstly, based on a two-phase stationary coordinate system, the voltage equation for the built-in permanent magnet synchronous motor (IPMSM) is established as follows: (1) (2) Combining formulas (1) and (2), we can obtain: (3) in, u α and u β This represents the components of the stator winding phase voltage in a two-phase stationary coordinate system. i α and i β This represents the components of the stator winding phase current in a two-phase stationary coordinate system. Lα and L β This represents the component of the stator winding phase inductance in a two-phase stationary coordinate system. L d and L q These represent the direct-axis inductance and quadrature-axis inductance of the stator winding, respectively. R This represents the phase resistance of the stator winding. oh e Indicates the rotor's electrical angular velocity. i e Indicates the rotor electrical angle, Indicates the magnetic flux linkage of the rotor permanent magnet. p Describes the differential operator. Indicates average inductance. Indicates half-differential inductance. Indicates coupled inductance. Represents the differential symbol. t This represents the differential time.

[0025] S2. Based on the modeling results, the flux linkage of the built-in permanent magnet synchronous motor (IPMSM) is estimated to obtain the motor flux linkage observed in real time by the online observer. In this embodiment, formula (3) is rewritten as the derivative form of the magnetic flux linkage: (4) in, Represents the stator voltage vector. Represents the stator current vector. T This indicates the transpose operation.

[0026] Then the motor flux l for: (5) in, Indicates average inductance. Indicates half-differential inductance. i e Indicates the rotor electrical angle, i α and i β This represents the components of the stator winding phase current in a two-phase stationary coordinate system. Indicates the magnetic flux linkage of the rotor permanent magnet. Represents the stator current vector. Represents the transformation matrix. This represents the product of transformation matrices.

[0027] in: (6) (7) Then the motor flux can be obtained. l for: (8) definition x for: (9) in, x Indicates magnetic flux. Indicates the cross-axis inductance of the stator winding. This represents the d-axis current.

[0028] Combining formula (4) and differentiating both sides of formula (9), we can obtain: (10) in, The derivative of the magnetic flux linkage. The derivative of the magnetic flux linkage of the motor. Describes the differential operator. Represents the stator current vector. Represents the stator voltage vector. R This represents the phase resistance of the stator winding. Indicates the cross-axis inductance of the stator winding. Indicates half-differential inductance. Represents the transformation matrix. Represents the d-axis current. oh e Indicates the rotor's electrical angular velocity. This represents the product of transformation matrices.

[0029] First, the cost function is defined using the Euclidean norm. h (· ,t ): (11) (12) The estimated value of the cost function is: (13) (14) in, This represents the observed value of the motor flux linkage. This represents the observed value of magnetic flux linkage.

[0030] The online flux linkage observer is constructed based on the above formula as follows: (15) (16) (17) in, The derivative of the magnetic flux linkage observation. The derivative of the observed flux linkage of the motor, Indicates the cross-axis inductance of the stator winding. Describes the differential operator. This indicates the gain of the flux observer.

[0031] (18) This is the error driving function for the online observer. c >0 represents the flux observer gain.

[0032] S3. Determine the polarity of the three-phase stator current to obtain the phase angle of the three-phase current, specifically: Define stator current vector i s The angle between the d-axis and the d-axis is i s ; According to the definition of the rotor dq synchronous rotating coordinate system, the d-axis and the rotor N pole position coincide. When the rotor position angle is... i e At that time, according to the inverse Park transform formula, we get: ;in, a Phase current i α Expressed by trigonometric function formulas; Based on the symmetrical distribution of the three-phase currents in the stator winding, the remaining two-phase currents are obtained. i b , i c phase angle d b and d c ,in, d a , d b and d c This represents the phase angle of the three-phase current.

[0033] In this embodiment, the stator current vector is defined. i s The angle between the d-axis and the d-axis is i s ,but: (19) in, and These represent the d-axis current and the q-axis current, respectively.

[0034] According to the definition of the rotor dq synchronous rotating coordinate system, the d-axis and the rotor N pole position coincide. When the rotor position angle is... i e At that time, according to the inverse Park transform formula, we can obtain a Phase current i α : (20) but a Phase current i α It can be expressed by trigonometric function formulas as follows: (twenty one) in, Represents the d-axis current. Indicates the rotor electrical angle, Represents the q-axis current. Represents the stator current vector. d a express a Phase current phase angle.

[0035] By combining formulas (19) and (21), we can obtain a Phase current phase angle d a for: (twenty two) From Clark's transformation formula, we know a Phase current i a and i α With the same phase, the remaining two phase currents can be obtained based on the symmetrical distribution of the three-phase currents in the stator windings. i b , i c phase angle d b and d c .

[0036] (twenty three) (twenty four) S4. Based on the motor flux linkage observed in real time by the online observer, the average error voltage compensation coefficient of the silicon carbide inverter is calculated. The average error voltage compensation coefficient of the silicon carbide inverter is a dynamically changing quantity, specifically: Based on the motor flux linkage observed in real time by the online observer, the estimated value of the rotor permanent magnet flux linkage is calculated, and the difference between the estimated value of the rotor permanent magnet flux linkage and the permanent magnet flux linkage constant is calculated. Based on the difference calculated, the average error voltage compensation coefficient of the silicon carbide inverter is obtained through the PI controller. kc .

[0037] In this embodiment, the commonly used inverter dead-zone compensation algorithm based on the average error voltage of silicon carbide inverters has the following compensation voltage formula: (25) in, T d Indicates dead time. T don This indicates a delay in service activation. T doff Indicates a shutdown delay. T s Indicates the PWM carrier period. V D This indicates the voltage drop across the anti-parallel diode. V ds This represents the voltage drop across the power device. U DC This represents the DC bus voltage. Due to the power devices of the silicon carbide inverter... T don , T doff , V ds , V d These parameters change with variations in load current, voltage, and temperature. Therefore, an average error voltage compensation coefficient for silicon carbide inverters is introduced. k c The coefficient k c With silicon carbide inverter power devices T don , T doff , V ds , V d It is related to parameters such as phase angle of three-phase current and is a dynamically changing quantity. d x This addresses the problem of inaccurate current polarity determination caused by repeated jumps at the zero-crossing point when the current decreases.

[0038] First, the estimated value of the rotor permanent magnet flux linkage is calculated using the motor flux linkage observed by the online observer, and the difference is taken from the permanent magnet flux linkage constant. Then, the average error voltage compensation coefficient of the inverter is calculated using the PI controller. k c Average error voltage compensation coefficient for silicon carbide inverters k c The calculation flowchart is as follows Figure 2 As shown, Figure 2 middle, This represents the initial value of the motor flux linkage observation. This represents the transformation matrix.

[0039] (26) in, This represents the average error voltage compensation coefficient for silicon carbide inverters. This represents the proportionality coefficient. Represents the integral coefficient. Represents the Laplace operator, Indicates the magnetic flux linkage of the rotor permanent magnet. This represents the estimated value of the rotor permanent magnet flux linkage.

[0040] S5. Based on the three-phase current phase angle and the average error voltage compensation coefficient of the silicon carbide inverter, the improved dead-zone compensation voltage of the three-phase online variable coefficient silicon carbide inverter is calculated, and the nonlinear dead-zone effect of the silicon carbide inverter is compensated by the improved dead-zone compensation voltage of the three-phase online variable coefficient silicon carbide inverter.

[0041] In this embodiment, the three-phase current phase angle is derived from formulas (20)-(24). d x Then, the dead-zone compensation voltage of the improved three-phase online variable coefficient silicon carbide inverter is obtained to compensate for the nonlinear dead-zone effect of the silicon carbide inverter. The improved dead-zone compensation voltage of the three-phase online variable coefficient silicon carbide inverter can be written as: (27) in, This indicates the dead-zone compensation voltage of the improved three-phase online variable coefficient silicon carbide inverter. This represents the average error voltage compensation coefficient for silicon carbide inverters. This indicates the switching frequency of the silicon carbide inverter. Indicates dead time. Indicates bus voltage. Represents a symbolic function. This indicates the phase angle of the three-phase current.

[0042] In this embodiment, the parameters of the permanent magnet synchronous motor (IPMSM) and silicon carbide (SiC) motor controller used in the experiment are shown in Table 1. Table 1 is a parameter table of the IPMSM motor and the silicon carbide (SiC) controller.

[0043] Table 1

[0044] In this embodiment, the switching frequency of the silicon carbide (SiC) motor controller is set to 30 kHz, the dead time is 2 μs, and Figure 3(a) is the experimental waveform of the motor phase current without dead time compensation. The time is 40 ms / div and the current is 2 A / div. Figure 3(b) is the Fourier analysis diagram of the motor phase current without dead time compensation. The total harmonic distortion (THD) of the phase current is 14.15%. In Figure 3(b), the horizontal axis represents the harmonic order and the vertical axis represents the percentage of harmonics relative to the fundamental frequency. Figure 4(a) shows the experimental waveform of the motor phase current based on the average error voltage dead zone compensation, with a time of 40 ms / div and a current of 2 A / div. Figure 4(b) shows the Fourier analysis of the motor phase current based on the average error voltage dead zone compensation. The total harmonic distortion (THD) of the phase current is 12.16%. The horizontal axis represents the harmonic order, and the vertical axis represents the percentage of harmonics relative to the fundamental frequency. Compared with the motor phase current without dead zone compensation, the THD is reduced by approximately 15%. Figure 5(a) shows the experimental waveform of the motor phase current using the dead zone compensation method proposed in this invention, with a time of 40 ms / div and a current of 2 A / div. Figure 5(b) shows the Fourier analysis of the motor phase current using the dead zone compensation method proposed in this invention. The total harmonic distortion (THD) of the phase current is 8.65%. The horizontal axis represents the harmonic order, and the vertical axis represents the percentage of harmonics relative to the fundamental frequency. Compared to the method without dead-zone compensation, the total harmonic distortion (THD) of the motor phase current is reduced by approximately 38%. Experimental results show that the THD of the motor phase current using the dead-zone compensation method proposed in this invention is the lowest, demonstrating that the proposed method is superior to commonly used dead-zone compensation methods based on average error voltage, and can effectively reduce the harmonic distortion of the motor phase current caused by the nonlinear dead-zone effect of silicon carbide inverters.

[0045] In summary, compared with existing technologies, the beneficial effects of this invention are: 1) It is not affected by the type of power devices in the silicon carbide inverter; both traditional silicon-based inverters and wide-bandgap silicon carbide inverters can be used. 2) It treats the silicon carbide inverter as a black box, independent of the nonlinear time-varying parameters of the silicon carbide inverter, solving the problem of requiring accurate silicon carbide inverter parameters in traditional dead-zone compensation methods based on average error voltage models. 3) It provides real-time online compensation for the nonlinear dead-zone effect of the silicon carbide inverter, with a high level of automation, eliminating the need for extensive offline testing and significantly reducing the workload of operators.

Claims

1. A method for compensating for nonlinear dead zone in an online variable coefficient silicon carbide inverter, characterized in that, Includes the following steps: Modeling of the built-in permanent magnet synchronous motor (IPMSM); Based on the modeling results, the flux linkage of the built-in permanent magnet synchronous motor (IPMSM) is estimated, and the motor flux linkage observed in real time by the online observer is obtained. The polarity of the three-phase stator current is determined to obtain the phase angle of the three-phase current; Based on the motor flux observed in real time by the online observer, the average error voltage compensation coefficient of the silicon carbide inverter is calculated, where the average error voltage compensation coefficient of the silicon carbide inverter is a dynamic variable. Based on the three-phase current phase angle and the average error voltage compensation coefficient of the silicon carbide inverter, the improved dead-zone compensation voltage of the three-phase online variable coefficient silicon carbide inverter is calculated, and the nonlinear dead-zone effect of the silicon carbide inverter is compensated by the improved dead-zone compensation voltage of the three-phase online variable coefficient silicon carbide inverter.

2. The online variable coefficient silicon carbide inverter nonlinear dead zone compensation method according to claim 1, characterized in that, The expression for modeling the built-in permanent magnet synchronous motor (IPMSM) is as follows: ； ； ； p= ； in, u α and u β This represents the components of the stator winding phase voltage in a two-phase stationary coordinate system. i α and i β This represents the components of the stator winding phase current in a two-phase stationary coordinate system. L α and L β This represents the component of the stator winding phase inductance in a two-phase stationary coordinate system. L d and L q These represent the direct-axis inductance and quadrature-axis inductance of the stator winding, respectively. R This represents the phase resistance of the stator winding. ω e Indicates the rotor's electric angular velocity. θ e Indicates the rotor electrical angle, Indicates the magnetic flux linkage of the rotor permanent magnet. p Describes the differential operator. Indicates average inductance. Indicates half-differential inductance. Indicates coupled inductance. Represents the differential symbol. t This represents the differential time.

3. The online variable coefficient silicon carbide inverter nonlinear dead zone compensation method according to claim 2, characterized in that, The expression for the motor flux linkage observed in real time by the online observer is: ； ； ； ； ； in, The derivative of the magnetic flux linkage observation. Represents the stator voltage vector. Represents the stator current vector. T This indicates the transpose operation. Indicates the flux linkage observer gain. Indicates the magnetic flux linkage observation value. Represents the cost function, The derivative of the observed flux linkage of the motor, This represents the observed value of the motor flux linkage. Represents the d-axis current. This represents the error driving function of the online observer.

4. The online variable coefficient silicon carbide inverter nonlinear dead zone compensation method according to claim 1, characterized in that, The polarity of the three-phase stator current is determined to obtain the phase angle of the three-phase current, which is as follows: Define stator current vector i s The angle between the d-axis and the d-axis is θ s ; According to the definition of the rotor dq synchronous rotating coordinate system, the d-axis and the rotor N pole position coincide. When the rotor position angle is... θ e At that time, according to the inverse Park transform formula, we get: ； in, a Phase current i α Expressed by trigonometric formulas: ； ； in, Represents the d-axis current. Indicates the rotor electrical angle, Represents the q-axis current. Represents the stator current vector. δ a express a Phase current phase angle. Based on the symmetrical distribution of the three-phase currents in the stator winding, the remaining two-phase currents are obtained. i b , i c phase angle δ b and δ c ,in, δ a , δ b and δ c This represents the phase angle of the three-phase current.

5. The online variable coefficient silicon carbide inverter nonlinear dead zone compensation method according to claim 6, characterized in that, Phase angle δ b and δ c The expression is as follows: ； 。 6. The online variable coefficient silicon carbide inverter nonlinear dead zone compensation method according to claim 1, characterized in that, The average error voltage compensation coefficient for silicon carbide inverters is calculated as follows: Based on the motor flux linkage observed in real time by the online observer, the estimated value of the rotor permanent magnet flux linkage is calculated, and the difference between the estimated value of the rotor permanent magnet flux linkage and the permanent magnet flux linkage constant is calculated. Based on the difference calculated, the average error voltage compensation coefficient of the silicon carbide inverter is obtained through the PI controller. k c .

7. The online variable coefficient silicon carbide inverter nonlinear dead zone compensation method according to claim 6, characterized in that, Average error voltage compensation coefficient of silicon carbide inverter k c The expression is as follows: ； in, Represents the proportionality coefficient. Represents the integral coefficient. Represents the Lagrangian operator, Indicates the magnetic flux linkage of the rotor permanent magnet. This represents the estimated value of the rotor permanent magnet flux linkage.

8. The online variable coefficient silicon carbide inverter nonlinear dead zone compensation method according to claim 1, characterized in that, The expression for the improved dead-zone compensation voltage of the three-phase online variable coefficient silicon carbide inverter is as follows: ； in, This indicates the dead-zone compensation voltage of the improved three-phase online variable coefficient silicon carbide inverter. This represents the average error voltage compensation coefficient for silicon carbide inverters. This indicates the switching frequency of the silicon carbide inverter. Indicates dead time. Indicates bus voltage. Represents a symbolic function. This indicates the phase angle of the three-phase current.

Images