Method for improving dead zone compensation based on Fourier expansion

Abstract

The invention provides an improved dead-time compensation method based on Fourier expansion, and relates to the technical field of inverter dead-time compensation, and the method comprises the steps: carrying out the sector division according to a current vector, carrying out the current vector detection of a three-phase current obtained through sampling, obtaining a current current vector, and determining a sector where the current current vector is located; calculating according to the sector to obtain compensation square wave voltage; fourier expansion is carried out on the compensation square wave voltage, and superposition is carried out on the compensation square wave voltage; and the superposed compensation voltage or compensation square wave voltage is used as the compensation voltage for improving dead-zone compensation and is used for dead-zone compensation. According to the method, improvement is carried out on the basis of a traditional mode that a sector is judged through a current vector angle, and then compensation square wave voltage is output, after Fourier expansion is carried out on an original square wave, the compensation waveform obtained after Fourier expansion is superposed, the problem that accurate compensation near a current zero crossing point is difficult is solved, and the compensation precision near the zero crossing point is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of inverter dead-time compensation, and particularly to an improved dead-time compensation method based on Fourier expansion. Background Art

[0002] Dead-time compensation technology is an important topic in the research of PWM inverters. In the vector control system of a servo system, SVPWM (Space Vector Pulse Width Modulation) technology is widely used. However, the dead-time in the inverter will cause problems such as a reduction in the fundamental component of the output voltage, distortion of the output current waveform, and pulsation of the output torque. These problems will reduce the control performance of the system. Therefore, it is particularly necessary to effectively compensate for the dead-time effect.

[0003] Currently, among the compensation methods for the dead-time effect, the current feedback type compensation method is a commonly used method. It mainly compensates the voltage by detecting the direction of the output three-phase current. The key to this method lies in accurately judging the zero-crossing point of the current. However, affected by the current clamping effect and system noise, etc., the zero-crossing point is often difficult to accurately judge. Specifically, when the current approaches the zero-crossing point (i.e., the sector switching boundary / edge), the current direction changes rapidly. Due to sampling delay or noise interference in the traditional method, it is difficult to accurately judge the sector belonging, resulting in inaccurate compensation voltage, causing current distortion or oscillation.

[0004] Based on this, the present invention is proposed. Summary of the Invention

[0005] The object of the present invention is to provide an improved dead-time compensation method based on Fourier expansion, to solve the problem that it is difficult to accurately compensate near the zero-crossing point of the current, and to improve the compensation accuracy near the zero-crossing point.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] An improved dead-time compensation method based on Fourier expansion, including: dividing sectors according to the current vector, detecting the current vector of the sampled three-phase current to obtain the current current vector, and determining the sector where the current current vector is located; calculating the compensated square-wave voltage according to the sector; performing Fourier expansion on the compensated square-wave voltage and superimposing it with the compensated square-wave voltage; the superimposed compensated voltage or compensated square-wave voltage is used as the compensated voltage for improving dead-time compensation for dead-time compensation.

[0008] The present invention provides a preferred solution. The method for improving dead-time compensation based on Fourier expansion further includes: dynamically dividing the compensation region according to the current vector, distinguishing the inside of the sector far from the zero-crossing point and the edge of the sector close to the zero-crossing point, and determining whether the current vector is located inside the sector or at the edge of the sector; when the current vector is located inside the sector, using the compensated square-wave voltage as the compensation voltage for improving dead-time compensation; when the current vector is located at the edge of the sector, using the superimposed compensated voltage as the compensation voltage for improving dead-time compensation.

[0009] The present invention provides a preferred solution. The edge of the sector is determined by the following steps: setting an angle threshold, and when the current vector is within the range of ± the threshold of the boundary of the sector where it is located, it is determined that the current vector is located at the edge of the sector.

[0010] The present invention provides a preferred solution. The Fourier expansion of the compensated square-wave voltage and the superposition with the compensated square-wave voltage specifically include: performing Fourier expansion on the compensated square-wave voltage, extracting harmonic components, and superimposing the harmonic components with the compensated square-wave voltage.

[0011] The present invention provides a preferred solution. For the extraction of harmonic components, specifically, the first and fifth Fourier harmonic components are extracted as the superimposed components.

[0012] Compared with the prior art, the present invention has the following beneficial technical effects:

[0013] The method for improving dead-time compensation based on Fourier expansion of the present invention is designed based on the current feedback type compensation method. It improves on the traditional method of judging the sector by the current vector angle and then outputting the compensated square-wave voltage. For the problem that it is difficult to accurately judge near the zero-crossing point, that is, at the edge of the sector, after performing Fourier expansion on the original square wave, the compensated waveform after Fourier expansion is superimposed, thereby improving the compensation accuracy, solving the problem that it is difficult to accurately compensate near the current zero-crossing point, and improving the compensation accuracy near the zero-crossing point. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the provided drawings.

[0015] Figure 1 It is the overall dead-time compensation block diagram under VF control;

[0016] Figure 2 It is the distribution diagram of the six voltage vectors of dead-time compensation in the plane;

[0017] Figure 3 Flow chart of the improved dead - zone compensation method based on Fourier expansion provided in Embodiment 1 of the present invention;

[0018] Figure 4 Flow chart of the improved dead - zone compensation method based on Fourier expansion provided in Embodiment 2 of the present invention;

[0019] Figure 5 Flow chart of introducing threshold determination in the improved dead - zone compensation method based on Fourier expansion provided in Embodiment 2 of the present invention;

[0020] Figure 6 Current waveform diagram before dead - zone compensation;

[0021] Figure 7 Current waveform diagram after adopting the improved dead - zone compensation method based on Fourier expansion provided in Embodiment 2 of the present invention. Specific embodiments

[0022] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0023] Please refer to Figure 1 , which gives a scheme for traditional dead - zone compensation based on the VF control method. In the VF control mode, the given voltage Us is obtained from the given frequency f. The voltage vector angle θ is obtained by integrating the given frequency. The alpha - axis and beta - axis voltages finally input to the SVPWM module can be obtained through the operations of sin and cos of θ. The SVPWM module outputs six - phase pulses to drive the on - off of the inverter (inverter bridge) to control the operation of the asynchronous motor. After sampling the three - phase current of the asynchronous motor by the sampler, it is input to the traditional dead - zone compensation module. Then, the three - phase current is first converted into the currents ialpha and ibeta in the two - phase stationary coordinate system, and then the corresponding current vector angle is obtained through the arctangent operation. According to the current vector angle, the sector where the voltage vector is located is judged, and thus the compensated voltages U αcomp and U βcomp in the two - phase stationary coordinate system are obtained. The sum of these voltages and the voltage input to the SVPWM module is the new input voltage.

[0024] The traditional current - feedback - type compensation method judges the sector (such as six - sector division) where it is located by detecting the current vector (amplitude and angle), and outputs a preset compensation square - wave voltage according to different sectors to cancel the voltage error caused by the dead - zone effect. Figure 1In traditional dead - zone compensation, after the three - phase current of the motor is sampled by a sampler, the current vector angle is calculated in the dead - zone compensation module. And according to the sector where the current vector angle is located, the voltage vectors to be compensated corresponding to different combinations of the three - phase currents are indirectly obtained. Then the compensated voltage vectors are respectively distributed to the alpha - axis and the beta - axis, and the corresponding alpha - axis compensation voltage U αcomp and beta - axis compensation voltage U βcomp are output. The magnitudes of them are respectively added to the alpha - axis and beta - axis voltages originally output to the SVPWM module through VF control as new voltages to input the SVPWM module. Please refer to Figure 2 . According to the current vector, the sector is divided. Specifically, it is divided into six sectors, with each sector being 60°. The six sectors are formed by the compensation voltage vectors (compensation square - wave voltages) corresponding to six different current polarity combinations. The compensation voltages are respectively distributed to the alpha - axis and the beta - axis, namely the corresponding valpha and vbeta, as shown in Table 1.

[0025] Table 1: Compensation voltage vectors corresponding to six different current polarity combinations

[0026]

[0027] Figure 2 In, U1 - U6 are the directions of the compensation voltage vectors corresponding to the six sectors. Taking U4 as an example, T d is the dead - zone time, T is the carrier period, and the calculation formula for the magnitude of the synthesized compensated voltage is: ΔU4 = U4 * 2T d / T. Similarly, for the magnitudes of the other six corresponding compensation voltages, they are evenly distributed in a circle with a difference of 60° in turn.

[0028] Sample the three - phase current (i a , i b , i c ) of the asynchronous motor, and then obtain the current vector angle. According to the six sectors where the current vector angle is located, the corresponding Valpha and vbeta to be compensated are obtained. Detect the current vector of the sampled three - phase current to obtain the current vector. The current vector includes the magnitude and the angle, and the current vector angle is specifically obtained through Clarke / Park transformation to get two voltage components Valpha and vbeta in the two - phase stationary coordinate system.

[0029] Next, the specific implementation scheme of the present invention based on the current - feedback - type compensation method, which is an improvement on the above - mentioned traditional method of judging the sector through the current vector angle and then outputting the compensation square - wave voltage, will be described in detail:

[0030] Example 1

[0031] Please refer to Figure 3 , a method for improving dead - zone compensation based on Fourier expansion provided in this embodiment is mainly implemented through the following steps:

[0032] S10. Divide sectors according to the current vector, detect the current vector of the three - phase current obtained by sampling, obtain the current current vector, and determine the sector where the current current vector is located.

[0033] S20. Calculate the compensation square - wave voltage according to the sector.

[0034] S30. Perform Fourier expansion on the compensation square - wave voltage and superimpose it with the compensation square - wave voltage. The superimposed compensation voltage is used as the compensation voltage for improving dead - zone compensation for dead - zone compensation.

[0035] The method for improving dead - zone compensation based on Fourier expansion provided in Embodiment 1 is designed based on the current - feedback compensation method. On the basis of the traditional method of judging the sector by the current - vector angle and then outputting the compensation square - wave voltage, for the problem that it is difficult to accurately judge near the zero - crossing point at the sector edge, after introducing Fourier expansion of the original square - wave, the compensation waveform after Fourier expansion is superimposed, so as to improve the compensation accuracy, solve the problem that it is difficult to accurately compensate near the current zero - crossing point, and improve the compensation accuracy near the zero - crossing point.

[0036] Embodiment 2

[0037] Please refer to Figure 4 , a method for improving dead - zone compensation based on Fourier expansion provided in this embodiment is mainly implemented through the following steps:

[0038] S10. Divide sectors according to the current vector, detect the current vector of the three - phase current obtained by sampling, obtain the current current vector, and determine the sector where the current current vector is located.

[0039] S20. Calculate the compensation square - wave voltage according to the sector.

[0040] S21. Dynamically divide the compensation area according to the current vector, distinguish the inside of the sector far from the zero - crossing point and the edge of the sector close to the zero - crossing point, and judge whether the current current vector is located inside the sector or at the edge of the sector.

[0041] S22. When the current current vector is located inside the sector, use the compensation square - wave voltage as the compensation voltage for improving dead - zone compensation.

[0042] S31. When the current current vector is located at the edge of the sector, perform Fourier expansion on the compensation square - wave voltage, extract the harmonic components, and superimpose the harmonic components with the compensation square - wave voltage. The superimposed compensation voltage is used as the compensation voltage for improving dead - zone compensation for dead - zone compensation.

[0043] The improved dead - zone compensation method based on Fourier expansion provided in Embodiment 2 is designed based on the current - feedback compensation method. It improves on the traditional method of judging the sector through the current vector angle and then outputting the compensated square - wave voltage. For the problem that it is difficult to accurately judge near the sector edge, i.e., near the zero - crossing point, after performing Fourier expansion on the original square wave, the interior of the sector far from the zero - crossing point and the sector edge close to the zero - crossing point are distinguished. It is judged whether the current vector is located inside the sector or at the sector edge. At the sector edge, the compensated waveform after Fourier expansion is superimposed, and the original square wave is retained inside the sector. That is, a zoning compensation strategy is adopted, thereby improving the compensation accuracy at the sector edge, solving the problem that it is difficult to accurately compensate near the current zero - crossing point, and improving the compensation accuracy near the zero - crossing point. At the same time, retaining the original square wave inside the sector can directly output the original square wave, avoiding complex calculations. Therefore, while maintaining the accuracy as a whole, the output efficiency can also be improved.

[0044] Please refer to Figure 5 , considering that misjudgment is likely to occur at the switching point of the sector edge, and the misjudgment at the sector - edge switching point is not simply a theoretical angle (such as 60°). Misjudgment often occurs at this theoretical edge angle (such as 58°, 59°, 61°, 62°, etc.). Therefore, the sector - edge switching point where misjudgment is likely to occur is actually an angle range. Therefore, considering this actual situation, the present invention provides a more preferred implementation manner, that is, setting a sector edge with a certain angle floating range to be closer to the actual sector - edge switching point, which is specifically determined through the following steps: setting an angle threshold φ, when the current vector (current - vector angle θ) is within the range of ± threshold φ of the boundary of the current sector, it is determined that the current vector is located at the sector edge. Specifically, it is judged whether the current - vector angle θ is within the range of ± threshold φ, calculating the voltage u1 of the superposition of the 1st and 5th Fourier harmonics as the compensation voltage, and calculating the compensated square - wave voltage u2 corresponding to the sector as the compensation voltage. In this preferred implementation manner, by superimposing the compensated waveform after Fourier expansion within the threshold of the sector - switching edge and retaining the original square wave for the rest, the occurrence of misjudgment phenomena can be further reduced, and the compensation accuracy can be further improved. Taking the compensated waveform after Fourier superposition to smooth the step - like transition within the original threshold to become an approximately linear transition can reduce the harmonic content and the compensation error. The value range of the threshold φ can be from 1° to 8°, preferably 5°. That is, within the range of ± 5° of the boundary of the sector, it is used as the sector - edge switching point. In an even more preferred implementation manner, for the extracted harmonic components, specifically extracting the 1st and 5th Fourier harmonic components as the superposition components has good effects while the calculation cost is low. Figure 5 In

[0045] Next, the experimental effect comparison is given to prove the effect of the method of the present invention. Experiments are carried out under the conditions of a carrier frequency of 10 kHz and a dead time of 4 microseconds. The images before compensation and after compensation by the method in the above-mentioned Embodiment 2 are as Figure 6 and Figure 7 shown. Figure 6 is the current waveform diagram before dead time compensation, that is, an asynchronous motor is used under the operating conditions of 50 Hz: power 30 kW, line voltage 380 V, frequency 50 Hz, stator resistance 0.143 Ω, stator leakage inductance 0.83 mH, rotor resistance 0.134 Ω, rotor leakage inductance 1.04 mH, mutual inductance 25.03 mH, moment of inertia 1.39 kg·m 2 The number of pole pairs is 4, the dead time is set to 4 microseconds, and the motor current waveform using VF control is adopted. It can be seen that there is an obvious current zero-crossing clamping effect, and the waveform distortion is relatively obvious. Figure 7 is the current waveform diagram after adopting the improved dead time compensation method based on Fourier expansion provided in Embodiment 2 of the present invention. The waveform obtained after the improved dead time method is relatively balanced, the current sinusoidality is good, and the motor operating state is stable.

[0046] The above-mentioned preferred embodiment solves the compensation defect of the traditional method near the zero-crossing point through Fourier expansion hybrid compensation and dynamic partitioning strategy, and finally realizes higher-precision dead time effect suppression, which is applicable to scenarios with strict requirements for current quality such as servo motors and grid-connected inverters.

[0047] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above-mentioned method embodiments. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0048] The technical features of the above-mentioned embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as the combinations of these technical features do not conflict, they should be considered as the scope described in this specification. And the above-mentioned embodiments only express several implementation manners of the present invention. Their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. For those of ordinary skill in the art, without departing from the concept of the present invention, several deformations and improvements can still be made, and these all belong to the protection scope of the present invention.

Claims

1. An improved dead zone compensation method based on Fourier expansion, characterized in that, Including: Dividing sectors according to the current vector, detecting the current vector of the three-phase current obtained by sampling to obtain the current current vector, and determining the sector where the current current vector is located; Calculating the compensation square-wave voltage according to the sector; Performing Fourier expansion on the compensation square-wave voltage and superimposing it with the compensation square-wave voltage; The superimposed compensation voltage or compensation square-wave voltage is used as the compensation voltage for improving dead-time compensation for dead-time compensation.

2. The improved dead zone compensation method based on Fourier expansion according to claim 1, wherein, Also including: Dynamically dividing the compensation region according to the current vector to distinguish the inside of the sector far from the zero-crossing point and the edge of the sector close to the zero-crossing point; When the current current vector is located inside the sector, the compensation square-wave voltage is used as the compensation voltage for improving dead-time compensation; when the current current vector is located at the sector edge, the superimposed compensation voltage is used as the compensation voltage for improving dead-time compensation.

3. The improved dead zone compensation method based on Fourier expansion according to claim 2, characterized in that, The sector edge is determined by the following steps: setting an angle threshold, and when the current current vector is within the range of ±threshold of the boundary of the sector where it is located, it is determined that the current current vector is located at the sector edge.

4. The improved dead zone compensation method based on Fourier expansion according to claim 1, characterized in that, The performing Fourier expansion on the compensation square-wave voltage and superimposing it with the compensation square-wave voltage specifically includes: performing Fourier expansion on the compensation square-wave voltage, extracting harmonic components, and superimposing the harmonic components with the compensation square-wave voltage.

5. The improved dead zone compensation method based on Fourier expansion according to claim 4, wherein The extracting harmonic components specifically extracts the 1st and 5th Fourier harmonic components as the superimposing components.

6. The method for improving dead zone compensation based on Fourier expansion according to claim 3, characterized in that, The value of the threshold is from 1° to 8°.

7. The improved dead zone compensation method based on Fourier expansion according to claim 1, characterized in that The detecting the current vector of the three-phase current obtained by sampling to obtain the current current vector, where the current vector includes amplitude and angle, and the current vector angle is specifically obtained by Clarke / Park transformation to obtain two voltage components Valpha and vbeta in the two-phase stationary coordinate system.

8. The method for improving dead zone compensation based on Fourier expansion according to claim 1, wherein, The dividing sectors according to the current vector specifically divides into six sectors, with each sector being 60°.

9. The improved dead zone compensation method based on Fourier expansion according to claim 8, characterized in that, The six sectors are formed by compensation voltage vectors corresponding to six different current polarity combinations.

Images