A hybrid multi-vector modulation (MPC) method

Abstract

The application provides a new mixed multi-vector modulation MPC method, and the steps are as follows: firstly, a reference voltage is calculated, and the position sector of the reference voltage is given; secondly, the basic voltage vector of the voltage source inverter in the position sector of the reference voltage is substituted into a cost function for calculation; then, four alternative voltage vector combinations are preselected according to the position sector of the reference voltage; finally, the values of the four alternative voltage vector combinations are calculated, the four alternative voltage vector combinations are substituted into the cost function for calculation, and the alternative voltage vector combination with the minimum cost function is screened out and applied to the control of the voltage source inverter in the next period. The application combines the traditional double-vector and triple-vector modulation MPC method, selects the optimal voltage vector from 18 alternating voltage vector combinations through simplification, reduces the calculation amount, and further improves the current control performance.

Description

Technical Field

[0001] This invention relates to the field of model predictive control technology for voltage source inverters, and in particular to a novel hybrid multi-vector modulation (MPC) method. Background Technology

[0002] Conventional modulation MPC strategies include single-vector MPC, dual-vector modulation MPC, and three-vector modulation MPC; single-vector MPC strategy: voltage source inverter topology such as Figure 1 As shown, where u dc Let L be the DC voltage, L be the filter inductance, and R be its parasitic resistance. a i b and i c It is a three-phase current, e a e b and e c This is the voltage of a three-phase power grid. For example... Figure 1 As shown, the voltage source inverter has 8 basic voltage vectors. They are u0(000), u1(100), u2(110), u3(010), u4(011), u5(001), u6(101), and u7(111), as follows. Figure 2 As shown, u0(000) and u7(111) are zero-voltage vectors with the same output voltage.

[0003] In the stationary αβ coordinate system, the mathematical model of the voltage source inverter can be expressed as (1).

[0004]

[0005] Where u αβ =[u α (k),u β (k)] T u α and u β The output voltage of the inverter in the αβ static coordinate system;

[0006] i αβ =[i α (k),i β (k)] T i α and i β e represents the output current of the inverter in the αβ static coordinate system. αβ =[e α (k),e β (k)] T e α and e β Let α be the grid voltage in the αβ static coordinate system.

[0007] Assuming the sampling period is T, after discretization, (2) can be obtained from (1).

[0008]

[0009] Where X(k) represents the variable X at the k-th instant.

[0010] Based on the deadbeat control principle, based on the reference current i αβref (k+2) The reference voltage is derived, and the reference voltage can be calculated using the linear interpolation theorem.

[0011]

[0012] Among them, u abref (k+1)=[u αref (k+1),u βref (k+1)] T u αref (k+1) and u βref (k+1) is the reference voltage.

[0013] To reduce the computational burden, a cost function based on the reference voltage can be defined to select the optimal voltage vector, as shown in (4).

[0014] G = |u αref (k+1)-u α (k+1)|+|u βref (k+1)-u β (k+1)| (4)

[0015] To obtain the optimal voltage vector, all eight basic voltage vectors are substituted into the cost function (4) to evaluate their voltage error. The voltage vector that minimizes the cost function is selected as the optimal voltage vector for the control of the inverter in the next stage.

[0016] Although the above-mentioned single-vector MPC strategy is simple to implement without using a PI controller and PWM algorithm, it has drawbacks such as large current ripple and large distortion.

[0017] Dual-vector modulation MPC strategy: To improve the control performance of single-vector MPC, dual-vector modulation MPC strategies have been proposed and studied by many researchers. In these methods, two voltage vectors are selected instead of one and applied in each control cycle.

[0018] The 12 voltage vector combinations can be defined based on the 8 basic voltage vectors of the voltage source inverter. They are u s1 (u0,u1),u s2 (u7,u2), u s3 (u0,u3),u s4 (u7,u4), us5 (u0,u5), u s6 (u7,u6), u s7 (u1,u2),u s8 (u2,u3), u s9 (u3,u4), u s10 (u4,u5), u s11 (u5,u6) and u s12 (u6,u1), such as Figure 3 As shown.

[0019] The relationship between the 12 voltage vector combinations and the 8 basic voltage vectors is shown in (5).

[0020]

[0021] Where t i,uj +t i,uk =T,t i,uj and t i,uk u j and u k The duration.

[0022] For example, u s1 The relationship between u0 and u1 can be expressed as follows:

[0023]

[0024] Based on (5), the values ​​of the 12 voltage vector combinations are determined, and the duration of each voltage vector is first calculated. According to the control principle of the modulation MPC strategy, it is assumed that the duration of each voltage vector is inversely proportional to its cost function value, as shown in (7).

[0025]

[0026] Where m is the normalization constant to be determined.

[0027] Therefore, the duration of each voltage vector can be calculated according to (7). For example, for the voltage vector combination u s1 It can be deduced that

[0028]

[0029] Among them G u0 and G u1 The value of the cost function shown in (4) can be obtained by substituting u0 and u1 into (4) respectively.

[0030] At the same time, consider t 1,u0 +t 1,u1 =T, which can be further derived

[0031]

[0032] For other voltage vector combinations, the duration can be calculated in the same way. Then, the values ​​of the 12 voltage vector combinations are derived according to (5). Finally, the 12 voltage vector combinations are substituted into (4) to obtain the optimal voltage vector combination. Since each voltage vector combination consists of two basic voltage vectors, dual-vector modulation MPC is achieved by applying one voltage vector combination in each control cycle.

[0033] Three-vector modulation MPC strategy: Although the two-vector modulation MPC strategy has better current control performance than the traditional single-vector MPC strategy, it still cannot minimize the control error. Therefore, three-vector modulation MPC strategies have been further proposed and studied by many researchers. In these methods, three voltage vectors are selected instead of two and applied in each control cycle.

[0034] like Figure 4 As shown, six voltage vector combinations are defined here, namely u t1 (u0,u1,u2), u t2 (u0,u2,u3), u t3 (u0,u3,u4), u t4 (u0,u4,u5), u t5 (u0,u5,u6), u t6 (u0,u6,u1).

[0035] The relationship between the 6 voltage vector combinations and the 8 basic voltage vectors is shown in (10).

[0036]

[0037] Where t i,ui +t i,uj +t i,uk =T,t i,ui t i,uj and t i,uk u i u j u k The duration.

[0038] Based on the principle of modulation MPC, the duration of each voltage vector can still be calculated according to the assumptions in (7).

[0039] For example, for voltage vector combination u t1 (11) can be derived from (7).

[0040]

[0041] Among them G u0 G u1 G u2 The value of the cost function shown in (4) can be obtained by substituting u0, u1, and u2 into (4) respectively.

[0042] For other voltage vector combinations, the duration can be calculated in the same way. Then, the values ​​of the six voltage vector combinations are derived according to (10). Finally, by substituting the six voltage vector combinations into (4), the optimal voltage vector combination can be selected for application. Thus, MPC with three-vector modulation is realized.

[0043] However, although many papers have studied the aforementioned two-vector and three-vector modulation MPC strategies and demonstrated their effectiveness through experimental results, few papers have conducted theoretical analyses to verify their theoretical effectiveness, which raises questions about their optimality. Summary of the Invention

[0044] Traditional MPC strategies suffer from large current ripple due to using only one voltage vector in each control cycle. While multi-vector modulation MPC strategies can reduce current ripple, they lack a theoretical basis and their optimality is uncertain. This invention proposes a novel hybrid multi-vector modulation MPC method to further improve the current control performance of voltage source inverters.

[0045] The technical solution of this invention is implemented as follows:

[0046] A novel hybrid multi-vector modulation (MPC) method comprises the following steps:

[0047] Step 1: Calculate the reference voltage and determine the sector where the reference voltage is located;

[0048] Step 2: Substitute the basic voltage vector of the voltage source inverter in the position sector of the reference voltage into the cost function for calculation;

[0049] Step 3: Based on the location sector of the reference voltage, pre-select four alternative voltage vector combinations;

[0050] Step 4: Calculate the values ​​of the four candidate voltage vector combinations, substitute the four candidate voltage vector combinations into the cost function for calculation, and select the candidate voltage vector combination with the smallest cost function to be applied to the control of the voltage source inverter in the next cycle.

[0051] Preferably, the reference voltage is calculated as follows:

[0052]

[0053] Where L is the filter inductance, R is the parasitic resistance, and u αβref(k+1) represents the reference voltage at time k+1 in the αβ static coordinate system, i αβref (k+2) represents the reference current at time k+2 in the αβ static coordinate system, i αβ (k+1) represents the reference current at time k+1 in the αβ static coordinate system, e αβ (k+1) represents the grid voltage at time k+1 in the αβ static coordinate system, and T represents the sampling period.

[0054] Preferably, the position sector of the reference voltage is shown in Table 1, where θ = arctan(u βref / u αref );

[0055] Table 1. Location sector of reference voltage

[0056]

[0057] Preferably, the expression for the cost function is:

[0058] G = |u αref (k+1)-u α (k+1)|+|u βref (k+1)-u β (k+1)|;

[0059] Among them, u αref (k+1) and u βref (k+1) are all reference voltages, u α (k+1) represents the inverter's output voltage along the α-axis in the αβ static coordinate system at time k+1, u β (k+1) represents the inverter's output voltage on the β axis at time k+1 in the αβ static coordinate system.

[0060] Preferably, the pre-selected candidate voltage vector combinations are shown in Table 2:

[0061] Table 2 Pre-selection method for hybrid voltage vector combination

[0062]

[0063] Preferably, the calculation method for the values ​​of the four candidate voltage vector combinations is as follows:

[0064]

[0065]

[0066]

[0067] Among them, t i,uj +t i,uk =T,ti,uj and t i,uk u j and u k The duration; m is the normalization constant to be determined; t i',ui +t i',uj +t i',uk =T,t i',ui t i',uj and t i',uk u i u j u k The duration of; i = 1, 2, ..., 12, i' = 1, 2, ..., 6.

[0068] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0069] 1) This invention introduces the principle of the proposed visualization analysis method, and on this basis verifies that the traditional multi-vector modulation MPC has better control performance than the single-vector modulation MPC;

[0070] 2) Based on the proposed visualization analysis method, this invention finds that traditional dual-vector and three-vector modulation MPC strategies cannot achieve optimal control. To this end, a new hybrid multi-vector modulation MPC method is proposed, which selects the optimal voltage vector from 18 alternating voltage vector combinations by simplifying the steps, thereby reducing the amount of computation and improving the current control performance. Attached Figure Description

[0071] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0072] Figure 1 This is the topology of the voltage source inverter of the present invention.

[0073] Figure 2 It is a combination of eight basic voltage vectors.

[0074] Figure 3 This refers to the voltage vector combination in the dual-vector modulation (MPC) method.

[0075] Figure 4 This refers to the voltage vector combination in the three-vector modulation MPC method.

[0076] Figure 5 This is a comparison of the errors between two-vector and three-vector vectors.

[0077] Figure 6This is a diagram showing the effectiveness region of the two-vector and three-vector modulation MPC methods.

[0078] Figure 7 This is the voltage vector combination of the present invention.

[0079] Figure 8 This is a comparison of the errors between the dual-vector and the basic voltage vector.

[0080] Figure 9 This is a comparison of the errors between the three-vector and the basic voltage vector.

[0081] Figure 10 For comparison of mixed multi-vector and dual-vector errors.

[0082] Figure 11 Comparison of mixed multi-vector and three-vector errors.

[0083] Figure 12 Photo of the experimental platform.

[0084] Figure 13 The current waveforms and FFT analysis results for four control methods with a reference current of 3A are shown; (a) single-vector MPC method; (b) dual-vector modulation MPC method; (c) three-vector modulation MPC method; and (d) multi-vector modulation MPC method.

[0085] Figure 14 The current waveforms and FFT analysis results for four control methods with a reference current of 8A are shown; (a) single-vector MPC method; (b) dual-vector modulation MPC method; (c) three-vector modulation MPC method; and (d) multi-vector modulation MPC method.

[0086] Figure 15 The current THDs for the four methods under different reference currents are shown.

[0087] Figure 16 Experimental results of the dynamic control performance of four methods are presented when the reference current is increased from 3A to 8A; (a) single-vector MPC method; (b) dual-vector modulation MPC method; (c) three-vector modulation MPC method; (d) multi-vector modulation MPC method. Detailed Implementation

[0088] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0089] To more clearly illustrate the differences between dual-vector and three-vector modulation MPC strategies, we will use rotation... Figure 5 Further drawing Figure 6 The effective regions of the two methods are illustrated, with the red area indicating that the dual-vector modulation MPC strategy is better, and the blue area indicating that the three-vector modulation MPC strategy is better.

[0090] from Figure 6 It is evident that the dual-vector modulation MPC strategy performs better when the reference voltage is located at the edge of each sector. However, the third-order vector modulation MPC strategy performs better when the reference voltage is located in the middle of each sector. Therefore, neither dual-vector nor third-vector modulation MPC strategies alone can achieve optimal control.

[0091] Based on the control principles of dual-vector and three-vector modulation MPC strategies, when the reference voltage is located at... Figure 6 When in the red zone, you should start from Figure 3 The 12 voltage vector combinations shown are u s1 -u s12 Select the optimal voltage vector. When the reference voltage is located at... Figure 6 When in the blue area, you should start from... Figure 4 The six voltage vector combinations shown are u t1 -u t6 Choose from among them. However, it is difficult to accurately distinguish between the red and blue areas. Therefore, to optimize control performance, it is necessary to compare... Figure 3 The 12 voltage vector combinations shown are u s1 -u s12 and Figure 4 The six voltage vector combinations shown are u t1 -u t6 The cost function is used to select the optimal voltage vector. This means that dual-vector and three-vector modulation MPC strategies should be combined. Therefore, this invention proposes a novel hybrid multi-vector modulation MPC strategy that combines traditional dual-vector and three-vector modulation MPC strategies. Its principle is as follows.

[0092] First, 18 voltage vector combinations are defined, including Figure 3 The 12 voltage vector combinations u are synthesized from two voltage vectors. s1 -u s12 ,as well as Figure 4 The six voltage vector combinations u are synthesized from three voltage vectors. t1 -u t6 The new 18 voltage vector combinations are as follows: Figure 7 As shown.

[0093] Secondly, the 12 voltage vector combinations u s1 -u s12 and 6 voltage vector combinations ut1 -u t6 The calculation methods are the same as those of traditional two-vector and three-vector modulation MPC strategies.

[0094] Third, for the proposed hybrid multi-vector modulation (MPC) strategy, 18 voltage vector combinations are first calculated in each control cycle. These are then substituted into the cost function (Equation 4), and the optimal voltage vector can be selected. Compared with traditional dual-vector and three-vector modulation (MPC) strategies, the proposed hybrid multi-vector modulation (MPC) strategy enriches the candidate voltage vectors and can further improve control performance.

[0095] To further demonstrate the effectiveness of the proposed hybrid multi-vector modulation (MPC) strategy, its effectiveness was verified based on the proposed visualization analysis method.

[0096] Still assuming reference voltage u αβref Located in sector I, then define two additional cost functions, error δ. 2h and δ 3h As shown in (20).

[0097]

[0098] and Figure 8 , Figure 9 and Figure 5 Similarly, according to (20), we can obtain Figure 10 and Figure 11 .

[0099] Depend on Figure 10 and Figure 11 It can be seen that δ 2h and δ 3h All values ​​are equal to or less than zero, indicating that the proposed hybrid multi-vector modulation (MPC) strategy has better control performance than traditional dual-vector and tri-vector modulation (MPC) strategies. This provides a solid theoretical foundation for the proposed hybrid multi-vector modulation (MPC) strategy.

[0100] The proposed novel hybrid multi-vector modulation (MPC) method significantly increases computational burden due to the need to evaluate 18 voltage vector combinations online in each control cycle. Therefore, a simplification was implemented to further reduce computational complexity. The simplified execution steps are shown below, where the number of AC voltage vector combinations to be evaluated online in each control cycle is reduced from 18 to 4.

[0101] Step 1: Calculate the reference voltage and determine the sector where the reference voltage is located;

[0102]

[0103] Where L is the filter inductance, R is the parasitic resistance, and uαβref (k+1) represents the reference voltage at time k+1 in the αβ static coordinate system, i αβref (k+2) represents the reference current at time k+2 in the αβ static coordinate system, i αβ (k+1) represents the reference current at time k+1 in the αβ static coordinate system, e αβ (k+1) represents the grid voltage at time k+1 in the αβ static coordinate system, and T represents the sampling period.

[0104] The location sector of the reference voltage is shown in Table 1, where θ = arctan(u βref / u αref );

[0105] Table 1. Location sector of reference voltage

[0106]

[0107] Step Two: Substitute the basic voltage vectors (i.e., the three basic voltage vectors) of the voltage source inverter within the reference voltage sector into the cost function for calculation, obtaining the minimum cost function value. The switching state corresponding to the minimum cost function is the optimal switching state for that reference voltage. Comparing the cost functions in Step Two and Step Three and visually analyzing the error function demonstrates the superiority of the invented method. For example, when the reference voltage position is sector I, only the cost functions of u0 (u7), u1, and u2 need to be calculated.

[0108] G = |u αref (k+1)-u α (k+1)|+|u βref (k+1)-u β (k+1)| (4)

[0109] Among them, u αref (k+1) and u βref (k+1) are all reference voltages, u α (k+1) represents the inverter's output voltage along the α-axis in the αβ static coordinate system at time k+1, u β (k+1) represents the inverter's output voltage along the α-axis in the αβ static coordinate system at time k+1.

[0110] Step 3: Calculate 18 voltage vector combinations within each control cycle. Then substitute them into equation (4) to obtain the optimal voltage vector. Based on the sector of the reference voltage, pre-select four alternative voltage vector combinations, as shown in Table 2. For example, when the reference voltage is located in sector I, the voltage vector combination u should be pre-selected. s1 u s2 u s7 and u t1 .

[0111] Table 2 Pre-selection method for hybrid voltage vector combination

[0112]

[0113] Step 4: Calculate the values ​​of the four candidate voltage vector combinations, and substitute the four candidate voltage vector combinations into the cost function (Equation 4) for calculation. Select the candidate voltage vector combination with the smallest cost function and apply it to the control of the voltage source inverter in the next cycle.

[0114] The calculation method for the values ​​of the four candidate voltage vector combinations is as follows:

[0115]

[0116]

[0117]

[0118] Among them, t i,uj +t i,uk =T,t i,uj and t i,uk u j and u k The duration; m is the normalization constant to be determined; t i',ui +t i',uj +t i',uk =T,t i',ui t i',uj and t i',uk u i u j u k The duration of; i = 1, 2, ..., 12, i' = 1, 2, ..., 6.

[0119] To further verify the effectiveness of the method of this invention and the proposed visualization analysis method, a system was established as follows: Figure 12 The experimental platform shown was used, and experimental research was conducted. The platform uses a DSP28335 digital signal processor as the main controller. A bidirectional DC power supply (APL-II) was used on the DC side. A three-phase programmable AC power supply (Ametek MX30) was used on the AC side. Experimental waveforms were recorded using a Yokogawa DLM4000 series oscilloscope. The parameters used in the experiment are shown in Table 3.

[0120] Table 3 System Parameters

[0121]

[0122] The steady-state and dynamic experimental results of four strategies—traditional single-vector modulation, dual-vector modulation, three-vector modulation, and hybrid multi-vector modulation—were compared.

[0123] First, to demonstrate the effectiveness of the proposed hybrid multi-vector modulation control strategy, its steady-state control performance was tested and compared with three other methods. Figure 13 and Figure 14 The figures show the current waveforms and FFT analysis results for four control methods when the reference currents are 3A and 8A, respectively.

[0124] from Figure 13 and Figure 14 It can be seen that both dual-vector and three-vector modulation MPC strategies exhibit better current control performance than the single-vector modulation MPC strategy. This is consistent with... Figure 8 and Figure 9 The theoretical analysis results shown are consistent, verifying the effectiveness of the visualization analysis method proposed in this invention. Furthermore, from... Figure 13 and Figure 14 It can also be seen that the proposed hybrid multi-vector modulation MPC strategy has better current control performance than the traditional dual-vector and three-vector modulation MPC strategies. This is consistent with... Figure 10 and Figure 11 The theoretical analysis results shown are consistent, which also verifies the effectiveness of the proposed hybrid multi-vector modulation (MPC) strategy and the proposed theoretical analysis method.

[0125] To further demonstrate the effectiveness of the proposed hybrid multi-vector modulation (MPC) strategy, the current THDs of the four methods under different reference currents were recorded and plotted, as follows: Figure 15 As shown. From Figure 15 As can be seen, compared with the other three methods, the method proposed in this invention has the lowest current THD when the reference current is changed. This further verifies the effectiveness of the method.

[0126] To further demonstrate the effectiveness of the proposed method, the dynamic control performance of the four methods was tested and compared. In this experiment, the reference current was increased from 3A to 8A. The results are as follows: Figure 16 As shown.

[0127] from Figure 16 It can also be seen that the method proposed in this invention has the best steady-state current control performance and the smallest current ripple compared with the other three methods.

[0128] In addition, from Figure 16 The results also show that the dynamic control performance of the four methods is similar. This verifies the effectiveness of the proposed hybrid multi-vector modulation (MPC) method and visualization analysis method.

[0129] Considering the significant current ripple inherent in traditional single-vector MPC strategies for voltage source inverters, and the lack of theoretical foundation for traditional dual-vector and three-vector modulation MPC strategies, this invention proposes a novel visualization analysis method, providing solid theoretical support for traditional dual-vector and three-vector modulation MPC strategies. However, based on the proposed visualization analysis method, it was found that the control performance of traditional multi-vector modulation MPC strategies is not optimal. Therefore, to improve control performance, this invention combines traditional dual-vector and three-vector modulation MPC strategies, further proposing a novel hybrid multi-vector modulation MPC method. The effectiveness of the hybrid multi-vector modulation MPC strategy is verified based on the proposed visualization analysis method. To reduce computational load, a simplified execution step for the MPC strategy is proposed. Finally, detailed steady-state and dynamic experimental results verify the effectiveness of the proposed hybrid multi-vector modulation MPC strategy and the effectiveness of the proposed visualization analysis method.

[0130] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A hybrid multi-vector modulation (MPC) method, characterized in that, The steps are as follows: Step 1: Calculate the reference voltage and determine the sector where the reference voltage is located; Step 2: Substitute the basic voltage vector of the voltage source inverter in the sector of the reference voltage location into the cost function to calculate the optimal voltage vector; Step 3: Based on the location sector of the reference voltage, pre-select four alternative voltage vector combinations based on the optimal voltage vector; The four pre-selected candidate voltage vector combinations are as follows: The preselected voltage vector corresponding to sector I is u. s1 ,u s2 ,u s7 ,u t1 The pre-selected voltage vector corresponding to sector II is u. s2 ,u s3 ,u s8 ,u t2 The pre-selected voltage vector corresponding to sector III is u. s3 ,u s4 ,u s9 ,u t3 The preselected voltage vector corresponding to sector IV is u. s4 ,u s5 ,u s10 ,u t4 The pre-selected voltage vector corresponding to sector V is u. s5 ,u s6 ,u s11 ,u t5 The pre-selected voltage vector corresponding to sector VI is u. s6 ,u s1 ,u s12 ,u t6 ; Step 4: Calculate the values ​​of the four candidate voltage vector combinations, substitute the four candidate voltage vector combinations into the cost function for calculation, and select the candidate voltage vector combination with the smallest cost function to be applied to the control of the voltage source inverter in the next cycle.

2. The hybrid multi-vector modulation (MPC) method according to claim 1, characterized in that, The reference voltage is calculated as follows: Where L is the filter inductance, R is the parasitic resistance, and u αβref (k+1) represents the reference voltage at time k+1 in the αβ static coordinate system, i αβref (k+2) represents the reference current at time k+2 in the αβ static coordinate system, i αβ (k+1) represents the reference current at time k+1 in the αβ static coordinate system, e αβ (k+1) represents the grid voltage at time k+1 in the αβ static coordinate system, and T represents the sampling period.

3. The hybrid multi-vector modulation (MPC) method according to claim 2, characterized in that, The reference voltage's position sector is as follows: Phase angle θ ranges from [0, π / 3), corresponding to sector I; phase angle θ ranges from [π / 3, π / 2), corresponding to sector II; phase angle θ ranges from [π / 2, π), corresponding to sector III; phase angle θ ranges from [π, π / 4), corresponding to sector IV; phase angle θ ranges from [π / 4, π / 5), corresponding to sector V; phase angle θ ranges from [π / 5, 2π), corresponding to sector VI; where phase angle θ = arctan(u βref / u αref ).

4. The hybrid multi-vector modulation (MPC) method according to claim 2 or 3, characterized in that, The expression for the cost function is: G=|u αref (k+1)-u α (k+1)|+|u βref (k+1)-u β (k+1)|; Among them, u αref (k+1) and u βref (k+1) are all reference voltages, u α (k+1) represents the inverter's output voltage along the α-axis in the αβ static coordinate system at time k+1, u β (k+1) represents the inverter's output voltage on the β axis at time k+1 in the αβ static coordinate system.

5. The hybrid multi-vector modulation (MPC) method according to any one of claims 1, 2, or 3, characterized in that, The calculation method for the values ​​of the four candidate voltage vector combinations is as follows: Among them, t i,uj +t i,uk =T,t i,uj and t i,uk u j and u k The duration; m is the normalization constant to be determined; t i',ui +t i',uj +t i',uk =T,t i',ui t i',uj and t i',uk u i u j u k The duration of; i = 1, 2, ..., 12, i' = 1, 2, ..., 6.

Images