SiC MOSFET inverter dead zone compensation method

Abstract

The invention relates to the technical field of semiconductor devices, and provides a SiC MOSFET inverter dead-time compensation method, which comprises the steps of S1, acquiring an output current of a SiC MOSFET inverter, and calculating a current ripple value when the SiC MOSFET inverter crosses a zero point; s2, a current zero-crossing region and a current non-zero-crossing region are divided according to the current ripple value, and when the current non-zero-crossing region is located, dead zone elimination is carried out on each phase of bridge arm of the SiC MOSFET; when the SiC MOSFET is in a current zero-crossing region, dead-zone compensation is carried out on each phase of bridge arm of the SiC MOSFET; and a software filter is arranged in the SiC MOSFET inverter to eliminate accidental harmonic waves in real time. According to the method provided by the invention, firstly, a mixed method of combining dead zone segmentation elimination and compensation is adopted, so that the dead zone effect is effectively inhibited, and the harmonic distortion rate of the current is well improved; on the second aspect, a software filter is arranged in the SiC MOSFET inverter, hardware cost is not increased, different filtering algorithms and parameters are selected according to requirements, and high flexibility and adjustability are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of semiconductor devices, and in particular to a dead zone compensation method for a SiC MOSFET inverter. Background Art

[0002] In order to avoid the problem of bridge arm shoot-through during the operation of the inverter, the necessary dead time needs to be inserted into the driving timing of the device. However, the dead time will bring problems such as inverter fundamental voltage loss and distortion. Especially when SiC MOSFET is used as a switching device, the higher switching frequency makes the waveform distortion more serious. The current dead time compensation methods can basically be divided into two types: one is the average voltage error compensation method, which has the advantage of being easy to implement, but the disadvantage is that the compensation is not accurate enough; the other is based on the pulse equivalent time compensation method, which can compensate the dead time more accurately, but the requirements for the control chip are also higher.

[0003] In order to avoid increasing hardware costs and to achieve dead-zone compensation more accurately, the dead-zone compensation strategy of the SiC MOSFET inverter needs to be improved, so as to improve the performance of the SiC MOSFET inverter without increasing hardware costs, and fully reflect the performance advantages that SiC MOSFET brings to the inverter. Summary of the invention

[0004] Based on the above situation, the main purpose of the present invention is to provide a dead zone compensation method for a SiC MOSFET inverter, so as to suppress the influence of the dead zone effect on the performance of SiC MOSFET as a switching device through a segmented hybrid method combining dead zone compensation and dead zone elimination.

[0005] The technical solution adopted by the present invention to solve the technical problem is as follows: The invention provides a SiC MOSFET inverter dead zone compensation method, comprising the steps of: S1, obtaining the output current of the SiC MOSFET inverter, and calculating the current ripple value of the SiC MOSFET inverter when the inverter passes through the zero point; S2, dividing the current zero-crossing region and the current non-zero-crossing region according to the current ripple value, when in the current non-zero-crossing region, performing dead zone elimination on each phase bridge arm of the SiC MOSFET; when in the current zero-crossing region, performing dead zone compensation on each phase bridge arm of the SiC MOSFET; the switching rule of dead zone elimination and dead zone compensation is: when the absolute value of the output current of the SiC MOSFET inverter is greater than the absolute value of a phase current, a single bridge arm action is adopted to eliminate the dead zone of the phase in the next PWM cycle; when the absolute value of the output current of the SiC MOSFET inverter is less than or equal to the absolute value of the phase current, a double bridge arm action is adopted to compensate the dead zone of the phase in the next PWM cycle.

[0006] Furthermore, in step S1, the specific method for obtaining the output current of the SiC MOSFET inverter is: according to the target current when the AC side load of the SiC MOSFET inverter is running and the measured power supply voltage, an instantaneous target current instruction is obtained, which is the output current of the SiC MOSFET inverter.

[0007] Furthermore, in step S1, the specific method for calculating the current ripple value of the SiC MOSFET inverter when it crosses the zero point is as follows: according to the fundamental amplitude and phase of the output current of the SiC MOSFET inverter, the amplitude and phase of the actual voltage fundamental of the output port of the SiC MOSFET inverter are calculated, and the precise threshold is calculated by combining it with the measured power supply voltage, which is the current ripple value of the SiC MOSFET inverter when it crosses the zero point.

[0008] Furthermore, the SiC MOSFET inverter is a three-phase two-level inverter. In step S1, the specific method for calculating the current ripple value is: obtaining the power supply voltage of the SiC MOSFET inverter and the phase voltages of the three phases, and when calculating the current ripple of one of the three phases, performing equivalent calculations on the other two phases using the Thevenin theorem to obtain the larger value of the absolute value of the current ripple amplitude of each switching cycle in the phase, and then obtaining the current ripple value at the current zero crossing point in a fundamental wave cycle.

[0009] Furthermore, a specific method for eliminating the dead zone is as follows: when the current is in the zero-crossing region, only one switch device on the phase bridge arm remains turned on, and the other switch device is turned off, and the turned-off switch device is freewheeled by a parallel diode.

[0010] Furthermore, a specific method for implementing dead zone compensation is: when the current is in a non-zero-crossing region, a dead zone time is added to the driving signal of the phase bridge arm, and dead zone compensation is performed by using an equivalent pulse exchange method.

[0011] Furthermore, the dead-time compensation method uses a built-in software filter in the SiC MOSFET inverter to eliminate accidental harmonics in real time.

[0012] Furthermore, the software filter adopts a software algorithm to add a subtraction channel and a filter on the basis of the second-order generalized integrator to form a double second-order generalized integrator.

[0013] The beneficial effects of the present invention include: In the first aspect, the present invention aims at the problem of current harmonic distortion caused by the dead zone effect in SiC MOSFET inverters, and proposes a hybrid method combining dead zone segment elimination and compensation. The compensation method applies two compensation methods according to whether the phase current passes through zero, effectively suppresses the dead zone effect, improves the harmonic distortion rate of the current, and improves the steady-state performance of the SiCMOSFET inverter. In the second aspect, the present invention has a built-in software filter formed by a dual second-order generalized integrator in the SiC MOSFET inverter, which does not increase the hardware cost, and realizes the selection of different filtering algorithms and parameters according to needs, with high flexibility and adjustability. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a step diagram of a SiC MOSFET inverter dead zone compensation method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods.

[0016] For SiC MOSFET inverters, since the actual switching action of the switching device requires a certain amount of time, in order to prevent the "shoot-through" phenomenon of the two switching devices on the same bridge arm, it is usually necessary to add a dead time between the complementary drive signals. The dead zone is a measure that must be taken to ensure the safe and reliable operation of the switching device, but its existence will cause the inverter output to produce fundamental voltage loss, increase of low-order harmonics, output current distortion and other dead zone effects. With the continuous increase in the switching frequency of modern devices, the impact of the dead zone effect on the output voltage and output current of the inverter is becoming more and more serious. Therefore, it is more important to eliminate and compensate the dead zone of the inverter. At present, the existing technology has proposed a large number of dead zone elimination and compensation methods. One type belongs to the dead zone compensation method, that is, disturbance, which applies various feedback or feedforward control algorithms to offset the dead zone effect. This type of method includes average voltage compensation method, predictive current control compensation method, adaptive control compensation method, disturbance observer compensation method. From the perspective of control, the harmonics generated by the dead zone are regarded as external. However, due to the nonlinear characteristics of the dead zone effect and the switching devices, it is difficult for this type of control compensation method to completely eliminate the harmonics caused by the dead zone. The other type belongs to the dead zone elimination method, that is, starting from the principle of dead zone generation, according to the different directions of the output inductor current, only one corresponding switching device in the same bridge arm is allowed to perform switching action, thereby avoiding the occurrence of the "shoot-through" phenomenon, and no longer needing to set the dead zone time. However, since the inductor current ripple is affected by the size of the inductor, and the current zero-crossing point detection using analog circuits or analog-to-digital converters (ADCs) will be affected by the sampling accuracy and sampling delay, the zero-crossing point detection used for current direction determination has a large ambiguity. Therefore, the theoretical inductor current zero-crossing point is actually an area with a certain width, and the detection and calculation of the zero-crossing area width are constrained by the accuracy of the corresponding algorithm. In this zero-crossing area, in order to allow the inverter to work normally, the strategy of restoring the drive signals of the upper and lower switching devices of the same bridge arm and re-joining the dead zone is generally adopted. The injection of the dead zone in the zero-crossing area of ​​the inductor current detection will cause the inverter output to produce obvious waveform distortion in this area. The characteristics and reliability of the simultaneous detection device greatly affect the effectiveness of this type of method.

[0017] Please refer to Figure 1In order to effectively suppress the dead zone effect of SiC MOSFET inverter, improve the harmonic distortion rate of current, and improve the steady-state performance of SiC MOSFET inverter, the present application provides a dead zone compensation method for SiC MOSFET inverter, adopts a hybrid method combining dead zone elimination and dead zone compensation, and develops a software algorithm filter algorithm at the same time. The method includes the steps of: S1, obtaining the output current of SiC MOSFET inverter, and calculating the current ripple value of SiC MOSFET inverter when it crosses zero point; S2, dividing the current zero-crossing area and the current non-zero-crossing area according to the current ripple value, when in the current non-zero-crossing area, performing dead zone elimination on each phase bridge arm of SiC MOSFET; when in the current zero-crossing area, performing dead zone compensation on each phase bridge arm of SiC MOSFET; the switching rule of dead zone elimination and dead zone compensation is: when the absolute value of the output current of SiC MOSFET inverter is greater than the absolute value of a phase current, a single bridge arm action is used to perform dead zone elimination on the phase in the next PWM cycle; when SiC When the absolute value of the output current of the MOSFET inverter is less than or equal to the absolute value of a phase current, a double-bridge arm action is used to implement dead zone compensation for the phase in the next PWM cycle.

[0018] Specifically, in order to avoid the problem of bridge arm shoot-through during the operation of the inverter, the necessary dead time needs to be inserted into the driving timing of the device. However, the dead time will bring problems such as inverter fundamental voltage loss and distortion. Especially when SiCMOSFET is used as a switching device, the higher switching frequency makes the waveform distortion more serious, which makes the traditional dead time compensation strategy applied to SiIGBT inverter no longer applicable. To this end, based on the traditional dead time elimination strategy, a segmented modulation dead time compensation strategy is proposed. This strategy obtains the current ripple value at the zero crossing point and divides the current zero crossing area and non-zero crossing area based on it. Affected by the inductor current ripple, the current zero crossing point is actually a zero crossing area with a certain width in theory. In order to enable the inverter to commutate safely and reliably in the zero crossing area, the dead time must be restored in the complementary switch tube. When the output current is in the non-zero-crossing area, each phase bridge arm only switches the effective devices, and the complementary devices are in the off state to increase the fundamental voltage amplitude; when in the zero-crossing area, the equivalent pulse compensation time is calculated for the errors caused by factors such as dead time and parasitic capacitance to compensate for the error voltage and reduce waveform distortion. The existence of current ripple requires the setting of the zero-crossing area near the current zero-crossing point, and the division of the current zero-crossing area depends on the ripple amplitude at the current zero-crossing point. If the ripple value of the inductor current is not accurate enough, the proportion of the single bridge arm action strategy will change, increase the harmonic content, and lead to the output voltage waveform distortion rate.

[0019] In order to accurately collect the output current of the SiC MOSFET inverter, in step S1, the specific method for obtaining the output current of the SiC MOSFET inverter is: according to the target current when the AC side load of the SiC MOSFET inverter is running and the measured power supply voltage, an instantaneous target current instruction is obtained, which is the output current of the SiC MOSFET inverter.

[0020] Specifically, fixed switching frequency current control is a real-time feedback control. It uses the error between the AC side current is(t) of the PWM rectifier and the target current isref (t) as the modulation wave, and uses a triangular wave or sawtooth wave with a fixed frequency as the carrier. After the two are compared and modulated, a PWM signal is generated to drive the power switch tube and control the input voltage of the AC side of the PWM rectifier. The physical meaning of fixed switching frequency control is relatively clear, the control structure is simple, and the switching frequency is fixed, which is convenient for calculating harmonics and designing filters. However, this method has a slow dynamic response speed when the switching frequency is low and is more sensitive to changes in system parameters. The above three current control methods can all be applied to the current control of AC electronic loads. Among them, the indirect current control has a slow dynamic response, and the system is easily disturbed and extremely unstable. It is rarely used in AC electronic loads today. Hysteresis current control and fixed switching frequency current control are both common control strategies for AC electronic loads. Since the switching frequency of hysteresis current is not fixed, it is impossible to quantitatively analyze the impact of the dead zone effect on the AC electronic load under hysteresis control. Fixed switching frequency control has a fixed switching frequency, which is easy to perform theoretical analysis and has a faster response than indirect current control, but it is more sensitive to fluctuations in system parameters. In order to better realize the load simulation function of AC electronic load and quantitatively analyze the dead zone effect of AC electronic load, a target current control strategy combining feedforward and closed loop is proposed.

[0021] In order to accurately determine the current ripple value, in step S1, the specific method for calculating the current ripple value of the SiC MOSFET inverter when it crosses the zero point is as follows: according to the fundamental amplitude and phase of the output current of the SiC MOSFET inverter, the amplitude and phase of the actual voltage fundamental of the output port of the SiC MOSFET inverter are calculated, and the precise threshold is calculated by combining it with the measured power supply voltage, which is the current ripple value of the SiC MOSFET inverter when it crosses the zero point.

[0022] Specifically, the existence of current ripple requires that a zero-crossing region must be set near the current zero-crossing point, and the division of the current zero-crossing region depends on the ripple amplitude at the current zero-crossing point. If the ripple value of the inductor current is not accurate enough, the proportion of the single bridge arm action strategy will change, the harmonic content will increase, and the output voltage waveform distortion rate will increase. In the zero-crossing region, the current directions on both sides of the zero-crossing point of a single switching cycle are opposite, and the actual output voltage can respond quickly at the moment when the corresponding switch tube is turned off, so that the actual and ideal output voltages in the zero-crossing region are consistent, and the dead zone effect is eliminated.

[0023] As another implementation scheme for calculating the current ripple value, the SiC MOSFET inverter is a three-phase two-level inverter. In step S1, the specific method for calculating the current ripple value is: obtaining the power supply voltage of the SiC MOSFET inverter and the phase voltages of the three phases, and when calculating the current ripple of one of the three phases, performing equivalent calculations on the other two phases using the Thevenin theorem to obtain the larger value of the absolute value of the current ripple amplitude of each switching cycle in the phase, and then obtaining the current ripple value at the current zero crossing point in a fundamental wave cycle.

[0024] Since the current detection accuracy and delay are affected by the detection means, the zero-crossing detection width cannot be exactly equal to the zero-crossing theoretical width, and the difference between the two is defined as the zero-crossing error width. The existence of the zero-crossing error width will cause the above-mentioned dead zone elimination method to be unable to completely eliminate the dead zone effect within the entire sinusoidal modulation cycle. If the detection width is greater than the theoretical width, the dead zone voltage is reintroduced within the zero-crossing error width, resulting in output voltage waveform distortion and the introduction of low-order harmonics; if the detection width is less than the theoretical width, the original current flow direction is destroyed within the zero-crossing error width, changing the normal PWM modulation mode, which also results in output voltage waveform distortion and the introduction of low-order harmonics.

[0025] In order to reduce the limitations of traditional dead zone elimination schemes, the specific method of implementing dead zone elimination in the present application is: when in the current zero-crossing region, only one switch device on the phase bridge arm remains turned on, and the other switch device is turned off, and the turned-off switch device is freewheeled by a parallel diode.

[0026] Specifically, the dead zone elimination is to selectively apply a switching signal to the switch device of the same bridge arm according to the change of the output current polarity, using the freewheeling effect of the anti-parallel diode, instead of applying complementary drive signals to the devices of the upper and lower bridge arms at the same time. According to the polarity of the current, the bridge arm can be equivalent to two conduction conditions. When the current is positive, the current will flow through the upper bridge arm channel and the lower bridge arm diode, and will not pass through the lower bridge arm channel and the upper bridge arm diode. Therefore, when the current is positive, the upper bridge arm switch tube is an effective device, and the lower bridge arm switch tube is an invalid device. Similarly, when the current is negative, the current will flow through the lower bridge arm channel and the upper bridge arm diode. At this time, the lower bridge arm switch tube is an effective device, and the upper bridge arm switch tube is an invalid device. Therefore, according to the polarity of the output current, only the effective SiC MOSFET device of the same bridge arm can be switched in each switching cycle, and the complementary bridge arm device is maintained at the off level, thereby eliminating the influence of the dead zone effect.

[0027] The existence of current ripple requires that a zero-crossing area must be set near the current zero-crossing point, and the division of the current zero-crossing area depends on the ripple amplitude at the current zero-crossing point. If the ripple value of the inductor current is not accurate enough, the proportion of the single bridge arm action strategy will change, the harmonic content will increase, and the output voltage waveform distortion rate will increase. The specific method for implementing dead zone compensation in this application is: when the current is in the non-zero-crossing area, the dead zone time is added to the drive signal of the phase bridge arm, and the dead zone compensation is performed by equivalent pulse exchange.

[0028] Traditionally, hardware filters are used to filter accidental harmonics, but the hardware filters are not flexible enough and not adaptable enough to the harmonic filtering parameters. The dead zone compensation method described in this application uses a built-in software filter in the SiC MOSFET inverter to eliminate accidental harmonics in real time.

[0029] Specifically, software filters are a digital signal processing technology that uses algorithms to weaken the impact of noise and improve the stability and accuracy of data. Applications of software filters include: Algorithm optimization and real-time improvement: With the increase in computing power, filtering algorithms are developing in a more efficient and real-time direction. For example, adaptive filtering algorithms can quickly adjust parameters in a changing environment to maintain the optimal filtering effect. Combination of deep learning and filtering algorithms: Deep learning has achieved remarkable results in image processing, speech recognition and other fields, and its application in filtering algorithms is becoming more and more extensive. Through deep neural networks, complex signals and noise can be modeled more accurately, thereby achieving more effective filtering. Development of multimodal filtering algorithms: In the fields of multi-sensor information fusion, image and video processing, multimodal filtering algorithms have been developed to meet complex data processing needs. Software-defined filtering: With the development of technologies such as software radio and software-defined networks, software-defined filtering has become a trend. By implementing filtering functions through software, filtering parameters can be flexibly adjusted to improve system adaptability and reliability. Deep learning filtering: Deep learning technology has achieved remarkable results in the fields of image and speech, and has also been gradually applied to the field of signal processing. Deep learning filters have the characteristics of adaptability and nonlinearity, which can achieve better filtering effects. With the continuous advancement of deep learning technology, deep learning filtering will be more widely used. Green environmental filtering: In the process of filter design and manufacturing, environmental factors are taken into consideration to reduce energy consumption and electromagnetic radiation. Green environmental filtering technology will become an important direction for the development of filters in the future. Integrated filters: With the advancement of integrated circuit technology, integrated filters have gradually become a trend. Integrated filters can achieve higher performance, smaller size and lower cost. In the future, integrated filters will be widely used in various application fields. These applications use software filters to play an important role in improving signal quality, suppressing noise, and signal separation.

[0030] A generalized second-order software filter is a very important filter in the field of signal processing. The orthogonal signal generator of the traditional single-phase second-order generalized integral model is used to generate orthogonal signals, which is easily affected by the DC and harmonic components in the power grid signal, resulting in a decrease in detection accuracy. This application improves the traditional single-phase second-order generalized integral model, while filtering out the DC component and maintaining the original high-frequency harmonic suppression advantages. The software filter uses a software algorithm to add a subtraction channel and a filter on the basis of the second-order generalized integrator to form a double second-order generalized integrator.

[0031] Specifically, the improved second-order generalized integrator adds a subtraction channel, and a filter is added to it. The purpose of adding the filter is to make it have greater attenuation in both low-frequency and high-frequency bands. By obtaining the characteristic transfer function with good attenuation characteristics in both low-frequency and high-frequency stages, the required filter transfer function characteristics can be inferred, thereby forming a double second-order generalized integrator. Compared with the traditional second-order generalized integrator control structure, the improved control model has greatly improved the performance of the traditional model under the disturbances of DC component, characteristic subharmonics and interharmonics, which can greatly improve the system's frequency locking ability and enhance the system's synchronous signal detection ability and anti-interference performance under non-ideal conditions of the power grid. The improved model has good anti-interference effect at both low and high frequencies.

[0032] It should be noted that the SiC MOSFET inverter dead zone compensation method proposed in this application will not change the duty cycle distribution of the effective switching device, and therefore will not affect the dynamic performance of the inverter output. On the other hand, the dead zone elimination method does not rely on high-precision current sampling devices, and can effectively eliminate the dead zone effect, reduce fundamental voltage loss, and significantly reduce low-order harmonic content and voltage THD, regardless of whether it is a resistive load or an inductive load, regardless of the size of the dead zone time. The voltage distortion suppression effect at the current zero crossing is particularly obvious, and the output sinusoidal waveform quality is greatly improved.

[0033] The SiC MOSFET inverter described in this application is not limited to single-phase or three-phase.

[0034] For those skilled in the art, various other corresponding changes and deformations can be made according to the technical solutions and concepts described above, and all of these changes and deformations should fall within the protection scope of the claims of the present invention.

[0035] It should be noted that the above-described embodiments are only some embodiments of the present invention, rather than all embodiments. The singular forms "a", "the" and "it" used in the embodiments and the appended claims are also intended to include plural forms, unless the context clearly indicates otherwise.

Claims

1. A SiC MOSFET inverter dead zone compensation method, characterized in that: The dead zone compensation method comprises the steps of: S1, obtain the output current of the SiC MOSFET inverter and calculate the current ripple value of the SiC MOSFET inverter when it crosses the zero point; S2, divide the current zero-crossing region and the current non-zero-crossing region according to the current ripple value. When in the current non-zero-crossing region, dead zone elimination is performed in each phase bridge arm of the SiC MOSFET; when in the current zero-crossing region, dead zone compensation is performed in each phase bridge arm of the SiC MOSFET; The switching rule of dead zone elimination and dead zone compensation is: when the absolute value of the output current of the SiC MOSFET inverter is greater than the absolute value of a phase current, a single bridge arm action is adopted for the phase to implement dead zone elimination in the next PWM cycle; when the absolute value of the output current of the SiC MOSFET inverter is less than or equal to the absolute value of a phase current, a double bridge arm action is adopted for the phase to implement dead zone compensation in the next PWM cycle.

2. The SiC MOSFET inverter dead zone compensation method according to claim 1, characterized in that: In step S1, the specific method for obtaining the output current of the SiC MOSFET inverter is: According to the target current and the measured power supply voltage when the AC side load of the SiC MOSFET inverter is running, the instantaneous target current instruction is obtained, which is the output current of the SiC MOSFET inverter.

3. The SiC MOSFET inverter dead zone compensation method according to claim 2, characterized in that: In step S1, the specific method for calculating the current ripple value of the SiC MOSFET inverter when it crosses the zero point is: According to the fundamental amplitude and phase of the output current of the SiC MOSFET inverter, the amplitude and phase of the actual fundamental voltage of the output port of the SiC MOSFET inverter are calculated, and the precise threshold is calculated by combining it with the measured power supply voltage, which is the current ripple value of the SiC MOSFET inverter when it crosses the zero point.

4. The SiC MOSFET inverter dead zone compensation method according to claim 1, characterized in that: The SiCMOSFET inverter is a three-phase two-level inverter. In step S1, the specific method for calculating the current ripple value is: The power supply voltage and three-phase phase voltages of the SiC MOSFET inverter are obtained. When calculating the current ripple of one of the three phases, the equivalent calculation is performed on the other two phases using the Thevenin theorem to obtain the larger absolute value of the current ripple amplitude in each switching cycle in the phase, and then the current ripple value at the current zero crossing point in a fundamental wave cycle is obtained.

5. The SiC MOSFET inverter dead zone compensation method according to claim 4, characterized in that: The specific method for implementing dead zone elimination is: when in the current zero-crossing region, only one switch device on the phase bridge arm remains turned on, and the other switch device is turned off, and the turned-off switch device is freewheeled by a parallel diode.

6. The SiC MOSFET inverter dead zone compensation method according to claim 4, characterized in that: The specific method for implementing the dead zone compensation is: when the current is in the non-zero-crossing area, the dead zone time is added to the driving signal of the phase bridge arm, and the dead zone compensation is performed by using an equivalent pulse exchange method.

7. The SiC MOSFET inverter dead zone compensation method according to claim 1, characterized in that: The dead-time compensation method uses a built-in software filter in the SiC MOSFET inverter to eliminate accidental harmonics in real time.

8. The SiC MOSFET inverter dead zone compensation method according to claim 7, characterized in that: The software filter adopts a software algorithm to add a subtraction channel and a filter on the basis of the second-order generalized integrator to form a double second-order generalized integrator.

Images