NPC type converter common-mode voltage suppression method with active midpoint potential control

Abstract

The invention discloses an NPC type converter common-mode voltage suppression method with active midpoint potential control, which adopts an improved virtual space vector pulse width modulation method to redefine a small vector and a middle vector, so that the maximum value of common-mode voltage generated by an obtained virtual voltage vector is Vdc / 6, and the maximum value of the common-mode voltage generated by the virtual voltage vector is Vdc / 6. Compared with a traditional virtual vector pulse width modulation method, the common-mode voltage value of Vdc / 3 is reduced by half, and therefore the suppression effect on the common-mode voltage is achieved. On the basis, the value of a virtual vector distribution coefficient k is reasonably selected, the problem of neutral-point potential imbalance caused by accumulation of neutral-point potential offset due to the initial voltage difference of upper and lower capacitors of the direct-current bus and the active current component effect is solved, the fluctuation of the neutral-point voltage is actively inhibited, and the neutral-point voltage balancing effect is improved. And the correctness and the effectiveness of the method are verified through an MATLAB / Simulink simulation test.

Description

Technical Field

[0001] The present invention belongs to the technical field of power electronics, and particularly relates to a common-mode voltage suppression method for an NPC type converter with active neutral point potential control. Technical Background

[0002] The neutral-point-clamped (NPC) three-level inverter has become a multi-level topology structure widely used in high-voltage and high-power applications due to its advantages such as low voltage stress on switching devices and low harmonic content in the output voltage. However, the NPC three-level inverter has problems of neutral point voltage fluctuation and common-mode voltage, which affect the safety and stability of its operation.

[0003] Regarding the neutral point voltage fluctuation problem of the NPC three-level inverter, many improved space vector modulation strategies have been proposed by domestic and foreign scholars. Some papers have theoretically proved that the neutral point voltage can be balanced by using the virtual space vector method. However, in practical applications, there are many factors affecting the neutral point voltage, such as the accumulation of neutral point potential offset caused by inconsistent capacitor parameters and the action of active current components, resulting in neutral point potential imbalance. Some scholars use the method of adjusting the action time of positive and negative small vectors to suppress the neutral point voltage fluctuation. However, at a high modulation ratio, the positive and negative small vectors at the same position cannot appear simultaneously, which limits the suppression effect of the neutral point voltage fluctuation. Later, different hybrid control strategies were proposed. According to different modulation ratios, when the reference vector is in different regions, the nearest three-vector method and the virtual space vector method are switched to control the neutral point potential balance. Most of the above control strategies do not involve the common-mode voltage suppression problem, and the above methods cannot control the situation where the neutral point potential originally has an offset or the offset caused by external changes. Therefore, it is necessary to break away from the traditional virtual space vector pulse width modulation method, redefine the small vectors and medium vectors, so that the common-mode voltage generated by the obtained virtual voltage vector is smaller, thereby achieving the suppression effect on the common-mode voltage. At the same time, under the condition of ensuring the neutral point voltage balance, through corresponding improvement measures, the fluctuation of the neutral point voltage is actively suppressed within a certain range of the output voltage. Summary of the Invention

[0004] In order to solve the problems of the prior art, the present invention provides a common-mode voltage suppression method for an NPC type converter with active neutral point potential control. This modulation method can balance the neutral point potential caused by external changes while suppressing the common-mode voltage generated by the three-level NPC converter.

[0005] To achieve the above object, the present invention designs a common-mode voltage suppression method for an NPC type converter with active neutral point potential control, which is characterized in that the method comprises the following steps:

[0006] Step 1) Obtain the three-phase modulation waves of the three-level NPC converter according to the carrier PWM modulation method;

[0007] Step 2) Synthesize the reference voltage vector V according to the three-phase modulation waves obtained in Step 1); ref ;

[0008] Step 3) Construct the space vectors of the three-level NPC converter. Each phase output of the three-level NPC converter has three states: P, O, and N, and a total of 27 switching states are output for the three phases, corresponding to 27 basic space vectors;

[0009] Step 4) Establish a space vector diagram in the α-β coordinate system with the amplitudes and phase angles of the basic space vectors corresponding to the three-phase switching states in Step 3). The space vector diagram of the three-level NPC converter consists of six large sectors; project the basic space vectors onto the α-β coordinate system, and the formula is:

[0010]

[0011] where θ is the angle between the reference voltage vector V ref and the α-axis in the α-β coordinate system;

[0012] The established space vectors Figure 1 There are a total of six large sectors, each large sector is 1 / 6 of the basic space vector diagram, and each 60° in the counterclockwise direction is a large sector;

[0013] Step 5) Synthesize virtual vectors from the basic space vectors corresponding to the three-phase switching states in Step 3) that generate a common-mode voltage amplitude of 0 or V dc / 6, and the selected vectors satisfy the condition that the average midpoint current is zero in principle within one switching period T S ;

[0014] Taking the synthesis method of the first large sector as an example, the formula for synthesizing virtual vectors is:

[0015]

[0016] where V VS1 , V VS2 are virtual small vectors, V VM is a virtual medium vector, V VL1 , V VL2 are virtual large vectors, and V V0 is a virtual zero vector;

[0017] Step 6) Introduce the distribution coefficient k. Taking the virtual medium vector synthesized in the first large sector as an example, redefine the virtual medium vector as:

[0018]

[0019] Determine the value of the distribution coefficient k according to the voltage difference between the upper and lower capacitors on the DC side;

[0020] Step 7) Calculate the action time of the virtual vector according to the volt-second balance principle;

[0021] Step 8) Calculate the action time of each basic vector; Substitute the action time of the virtual vector into the virtual vector synthesis formula, and after inverse solution, obtain the action time of the basic vector, and determine the switching state output sequence of the basic vector;

[0022] 9) Control the switching states of the switches in each item of the three-level NPC converter according to the switching state output sequence corresponding to the basic vectors in all sectors, and complete the common-mode voltage suppression of the NPC converter with active neutral point potential control.

[0023] Compared with the prior art, the beneficial effects of the present invention are as follows: The common-mode voltage suppression method of the NPC converter with active neutral point potential control designed by the present invention adopts an improved virtual space vector pulse width modulation method, redefines the small vectors and medium vectors, so that the maximum value of the common-mode voltage generated by the obtained virtual voltage vector is V dc / 6, which is half of the common-mode voltage value of V dc / 3 generated by the traditional virtual vector pulse width modulation method, thus achieving the effect of suppressing the common-mode voltage. At the same time, by reasonably selecting the value of the virtual medium vector distribution coefficient k, the problem of midpoint potential imbalance caused by the accumulation of the initial voltage difference between the upper and lower capacitors of the DC bus and the midpoint potential offset caused by the action of the active current component is solved, and the fluctuation of the midpoint voltage is actively suppressed. Description of the Drawings

[0024] Figure 1 It is the topological structure diagram of the three-level NPC converter.

[0025] Figure 2 It is the basic space vector diagram of the three-level NPC converter.

[0026] Figure 3 It is the virtual vector space vector diagram synthesized in the first large sector of the common-mode voltage suppression method of the NPC converter with active neutral point potential control of the present invention.

[0027] Figure 4 It is the small sector division space vector diagram of the first large sector of the common-mode voltage suppression method of the NPC converter with active neutral point potential control of the present invention.

[0028] Figure 5 It is the common-mode voltage diagram of an embodiment of the common-mode voltage suppression method of the NPC converter with active neutral point potential control of the present invention.

[0029] Figure 6This is the capacitor voltage waveform diagram of an embodiment of the common-mode voltage suppression method for an NPC converter with active neutral point potential control according to the present invention. Detailed implementation manners

[0030] The present invention will be further described in detail below with reference to the accompanying drawings.

[0031] The topology of a three-level NPC converter is as Figure 1 shown. The converter has three-phase bridge arms, and each phase bridge arm has four power switching tubes S a1 -S a4 and two clamping diodes D a1 -D a2 which are composed. The DC side consists of two capacitors C P and C N . The midpoint O of the two capacitors is connected to the clamping diodes to achieve midpoint clamping. V dc is the DC side bus voltage, and E a , E b , E c are the three-phase voltages of the power grid.

[0032] The present invention provides a common-mode voltage suppression method for an NPC converter with active neutral point potential control, and this method includes the following steps:

[0033] Step 1) Obtain the three-phase modulation waves of the three-level NPC converter according to the carrier PWM modulation method;

[0034] The three-phase modulation waves of the three-level NPC converter are:

[0035]

[0036] where m is the modulation degree, and its value is in the range of (0 to 1), ω is the fundamental angular frequency, t is the time, U a is the phase A voltage, U b is the phase B voltage, and U c is the phase C voltage;

[0037] Step 2) Synthesize the reference voltage vector V ref .

[0038] The synthesis formula of the reference voltage vector V ref is:

[0039]

[0040] where,

[0041] Step 3) Construct the space vectors of the three-level NPC converter. Each phase output of the three-level NPC converter has three states: P, O, and N. There are 27 switching states in total for the three-phase output, corresponding to 27 basic space vectors.

[0042] The space vector expression V of the three-level NPC converter is:

[0043]

[0044] Where S a 、S b and S c are the switching states of phase A, phase B, and phase C of the three-level NPC converter respectively. S a 、S b 、S c ∈(-1, 0, 1), and the switching states P, O, and N are represented by 1, 0, and -1 respectively. According to the magnitude of the basic space vectors, they are divided into basic zero vectors, basic small vectors, basic medium vectors, and basic large vectors. Among them, there are 3 basic zero vectors, 12 basic small vectors, 6 basic medium vectors, and 6 basic large vectors.

[0045] Step 4) In the α-β coordinate system, establish a space vector diagram with the magnitude and phase angle of the basic space vectors corresponding to the three-phase switching states in Step 3). The space vector diagram of the three-level NPC converter consists of 6 large sectors. The formula for projecting the basic space vectors onto the α-β coordinate system is:

[0046]

[0047] Where θ is the angle between the reference voltage vector V ref and the α-axis in the α-β coordinate system.

[0048] The established space vector Figure 1 has a total of six large sectors. Each large sector is 1 / 6 of the basic space vector diagram, and each 60° counterclockwise is a large sector, namely I-VI, as Figure 2 shown.

[0049] Step 5) Synthesize virtual vectors from the basic space vectors corresponding to the three-phase switching states in Step 3) that generate a common-mode voltage amplitude of 0 or V dc / 6, and the selected vectors satisfy the principle that the average midpoint current is zero within one switching period T S . First, perform the synthesis of virtual vectors in the first large sector, and transfer the reference voltage vector to the first large sector of the three-level NPC converter space vector diagram through the rotation transformation method for other sectors.

[0050] The formula for the common-mode voltage V com generated by 27 space vectors is:

[0051]

[0052] Table 1 shows the common-mode voltages corresponding to the 27 voltage vectors of the three-level NPC converter.

[0053] Table 1 Common-mode voltages of the three-level NPC converter

[0054]

[0055]

[0056] The formula for judging the large sector position is:

[0057]

[0058] Among them, the ceil() function represents rounding up. That is, the first large sector corresponds to θ ∈ (0 to 60°), the second large sector corresponds to θ ∈ (60° to 120°), the third large sector corresponds to θ ∈ (120° to 180°), the fourth large sector corresponds to θ ∈ (180° to 240°), the fifth large sector corresponds to θ ∈ (240° to 300°), and the sixth large sector corresponds to θ ∈ (300° to 360°).

[0059] The rotation transformation formula for projecting other sectors onto the first sector is:

[0060] S = θ - 60°(N - 1) (7)

[0061] Among them, S represents the position of the projected reference voltage vector V ref corresponding to the first large sector.

[0062] In addition, according to the following formula, analyze the influence of the basic voltage vector on the midpoint current:

[0063] i np = (1 - |S a |)i a + (1 - |S b |)i b + (1 - |S c |)i c (8)

[0064] Among them, i a 、i b 、i c are the three-phase currents of the converter.

[0065] According to the analysis of Equation (8), it can be known that the midpoint current i np generated by the zero vector and the large vector is 0, and has no influence on the midpoint potential in principle. Table 2 shows the midpoint currents generated by the small vector and the medium vector.

[0066] Table 2 Midpoint current generated by small vectors and medium vectors

[0067]

[0068]

[0069] Taking the first large sector I synthesis method as an example, the synthesized virtual vectors are as follows Figure 3 shown, and the synthesis formula is

[0070]

[0071] where V VS1 , V VS2 are virtual small vectors, V VM is a virtual medium vector, V VL1 , V VL2 are virtual large vectors, V V0 is a virtual zero vector

[0072] According to the synthesized virtual vectors, each large sector is further divided into 5 small sectors. Taking the first large sector as an example, as Figure 4 shown, the small sectors are divided by the following formula

[0073]

[0074] where L1 - L5 are the regional division lines

[0075] In the first small sector of the first large sector, the midpoint current generated by the basic vectors V OOO , V PNN , V PPN is zero, and the midpoint current generated by the basic vectors V OPN , V PON , V PNO is i a , i b , i c . In one cycle, the midpoint current generated by the virtual vectors V V0 , V VL1 , V VL2 , V VS1 , V VS2 is 0. Also, since the sum of the three-phase currents output by the three-phase converter within one switching cycle T S is 0, that is, i a +i b +i c =0, then the midpoint current i vm generated by the virtual medium vector = (i a +i b +i c ) = 0

[0076] Therefore, the maximum value of the common-mode voltage amplitude of all defined virtual vectors is ±V dc / 6, and it will not affect the midpoint potential in terms of the synthesis principle.

[0077] Step 6) Introduce the distribution coefficient k. Taking the virtual middle vector synthesized in the first largest sector as an example, redefine the virtual middle vector as:

[0078]

[0079] At this time, the virtual middle vector V VM generates a midpoint current i np which is:

[0080] i np = ki a + (1 - 2k)i b + ki b = (1 - 3k)i b (12)

[0081] From this, the charge Q flowing out of the capacitor midpoint under the action of the virtual middle vector is obtained as:

[0082]

[0083] When k = 1 / 3, the charge Q flowing out of the midpoint is Q = 0, and at this time, the midpoint current generated by the virtual middle vector is 0.

[0084] When k ∈ (0 to 1 / 3), i b > 0, the charge Q flowing out of the midpoint is Q > 0, then U CP becomes smaller, U CN becomes larger, when i b < 0, the charge Q flowing out of the midpoint is Q < 0, then U CP becomes larger, U CN becomes smaller.

[0085] When k ∈ (1 / 3 to 1 / 2), i b > 0, the charge Q flowing out of the midpoint is Q < 0, then U CP becomes larger, U CN becomes smaller, when i b < 0, the charge Q flowing out of the midpoint is Q > 0, then U CP becomes smaller, U CN becomes larger.

[0086] Among them, U CP、 U CN are the voltages of the upper and lower capacitors C P and C N on the DC side of the three-level NPC converter.

[0087] Set A as an allowable voltage deviation value, and let ΔV be the voltage difference between the upper and lower capacitors on the DC side, that is, ΔV = U CP -U CN , the maximum value of ΔV caused by the virtual middle vector is ΔV max . Divide the value range of k into two segments of (0~1 / 3) and (1 / 3~1 / 2) with 1 / 3 as the boundary, and set k1∈(0~1 / 3), k2∈(1 / 3~1 / 2). By reasonably using the virtual middle vector distribution coefficient k, ΔV can be restricted within the set voltage fluctuation range, that is, ΔV max < A. The specific adjustment principle of k is as follows:

[0088] (1) When -A ≤ ΔV ≤ A, it indicates that the midpoint voltage fluctuation is within the allowable range, and at this time, there is no need to suppress the midpoint voltage fluctuation.

[0089] (2) When ΔV > A, it indicates that the midpoint voltage fluctuation value exceeds the allowable range. At this time, it is necessary to make U CP become smaller and U CN become larger, then it is necessary to make the charge Q flowing out of the midpoint < 0. It can be seen from Equation (13) that when i b < 0, let k = k1; when i b > 0, let k = k2, which can make the charge Q < 0.

[0090] (3) When ΔV < -A, it indicates that the midpoint voltage fluctuation value exceeds the allowable range. At this time, it is necessary to make U CP become larger and U CN become smaller, then it is necessary to make the charge Q flowing out of the midpoint > 0. Also, it can be seen from Equation (13) that when i b < 0, let k = k2; when i b > 0, let k = k1, which can make the charge Q > 0.

[0091] Therefore, by adjusting the distribution coefficient k, the situation where the midpoint voltage is originally offset can be controlled.

[0092] Step 7) According to the volt-second balance principle, calculate the action time of the virtual vector; from formula (14), taking the third smallest sector in the first large sector as an example, calculate the action time T VS1 、T VL1 、T VM .

[0093]

[0094] Among them, T S is the action time of the reference voltage vector V ref ;

[0095] To simplify the calculation, the present invention selects as Figure 4The g-h coordinate system shown is used to calculate the action time of each virtual vector. The formula for converting the α-β coordinate system to the g-h coordinate system is:

[0096]

[0097] Taking the third smallest sector of the first large sector as an example, Equation (14) is expressed in the g-h coordinate system as:

[0098]

[0099] where V VM(g) and V VM(h) are the components of the middle vector V VM on the g and h axes respectively.

[0100] The action time of the virtual vector can be calculated from Equation (16):

[0101]

[0102] Similarly, the action time of the virtual vector when the reference voltage vector V ref is located in other small sectors of the first large sector can be calculated. As shown in Table 3, the action time of the virtual vector in other large sectors can be obtained similarly by rotating and transforming to the first large sector.

[0103] Table 3 Action time of virtual vectors in each small sector of the first large sector

[0104]

[0105] Step 8) Calculate the action time of each basic vector; substitute the action time of the virtual vector into the virtual vector synthesis formula, and after inverse solution, obtain the action time of the basic vector and determine the switching state output sequence of the basic vector. The principle for determining the switching sequence of each phase switch is that the vectors between sectors should be smoothly switched, that is, the switching state needs to follow the conversion between P and O or N and O, and on the premise of ensuring a low switching frequency as much as possible, determine the reasonable switching sequence of the middle vector in each small sector. Table 4 shows the state sequence corresponding to the vector action in the first large sector.

[0106] Table 4 State sequence corresponding to the vector action in the first large sector

[0107]

[0108] Step 9) Control the switching states of each switch in each phase on the three-level NPC type converter according to the switching state output sequence corresponding to all sector basic vectors, that is, the P state corresponds to the switches S a1 and S a2 being turned on, S a3、 and S a4 being turned off, and the O state corresponds to the switches S a2 and Sa3 Turn on, S a1 and S a4 Turn off, the N state corresponds to the switch S a3、 S a4 Turn on, S a1 and S a2 Turn off, complete the common - mode voltage suppression of the NPC - type converter with active neutral - point potential control.

[0109] To verify the effectiveness of the proposed method for common - mode voltage suppression of an NPC - type converter with active neutral - point potential control based on virtual - vector modulation, a simulation analysis is carried out in MATLAB / Simlink, and the simulation parameters are shown in Table 5.

[0110] Table 5 Simulation parameters

[0111]

[0112] Using MATLAB / Simlink software, according to the key parameters in Table 5, the simulation results show that the line voltage, phase voltage, and phase current output by the method of the present invention all present relatively ideal sine waves, and the total harmonic distortion THD of the phase current is within the allowable range of 5%. In terms of common - mode voltage suppression, the amplitude of the common - mode voltage generated by the method proposed in the present invention is ±125V (±V dc / 6), as Figure 5 shown. Compared with the common - mode voltage of the traditional virtual - vector pulse - width modulation method, it is reduced by half, and can effectively suppress the common - mode voltage generated by the NPC three - level converter. Given an initial potential difference between the upper and lower capacitors, the initial voltage of the upper capacitor U CP is set to 380V, and the initial voltage of the lower capacitor U CN is set to 370V, that is, simulating the situation where the neutral - point potential originally has an offset. The method can actively control the neutral - point potential ΔV within the range of 0.3V in 0.12s, as Figure 6 shown. Therefore, the method for common - mode voltage suppression of an NPC - type converter with active neutral - point potential control provided by the present invention can not only effectively suppress the common - mode voltage of the NPC three - level inverter, but also actively balance the situation where the neutral - point potential originally has an offset.

[0113] Matters not described in the present invention are applicable to the prior art.

Claims

1. A method for suppressing the common - mode voltage of an NPC - type converter with active neutral - point potential control, characterized in that, The method includes the following steps: Step 1) Obtain the three-phase modulation waves of the three-level NPC converter according to the carrier PWM modulation method; Step 2) Synthesize the reference voltage vector V according to the three-phase modulation waves obtained in Step 1) ref ; Step 3) Construct the space vectors of the three-level NPC converter. Each phase output of the three-level NPC converter has three states: P, O, and N, and a total of 27 switching states are output for the three phases, corresponding to 27 basic space vectors; Step 4) Establish a space vector diagram in the α-β coordinate system based on the amplitudes and phase angles of the basic space vectors corresponding to the three-phase switching states in Step 3). The space vector diagram of the three-level NPC converter consists of six large sectors; project the basic space vectors onto the α-β coordinate system, and the formula is: where θ is the angle between the reference voltage vector V ref and the α-axis in the α-β coordinate system; The established space vector diagram has a total of six large sectors. Each large sector is 1 / 6 of the basic space vector diagram, and each 60° in the counterclockwise direction is a large sector; Step 5) Synthesize virtual vectors from the basic space vectors corresponding to the three-phase switch states in Step 3) that generate common-mode voltage amplitudes of 0 or V dc / 6, and the selected vectors satisfy the principle that the average midpoint current is zero within one switching period T S ; Taking the synthesis method of the first large sector as an example, the formula for synthesizing the virtual vector is: Among them, V VS1 , V VS2 are virtual small vectors, V VM is a virtual medium vector, V VL1 , V VL2 are virtual large vectors, V V0 is a virtual zero vector; Step 6) Introduce the distribution coefficient k. Taking the virtual middle vector synthesized in the first large sector as an example, redefine the virtual middle vector as: Determine the value of the distribution coefficient k according to the voltage difference between the upper and lower two capacitors on the DC side; Step 7) Calculate the action time of the virtual vector according to the volt-second balance principle; Step 8) Calculate the action time of each basic vector; substitute the action time of the virtual vector into the virtual vector synthesis formula, and after inverse solution, obtain the action time of the basic vector, and determine the switching state output sequence of the basic vector; 9) Control the switching states of each switch in each item of the three-level NPC type converter according to the switching state output sequences corresponding to the basic vectors in all sectors, and complete the common-mode voltage suppression of the NPC type converter with active neutral point potential control.

2. The method for suppressing the common-mode voltage of an NPC-type converter with active neutral-point potential control according to claim 1, characterized in that In Step 1), the three-phase modulation waves of the three-level NPC converter are: where m is the modulation index with a range of 0 to 1, ω is the fundamental angular frequency, t is time, U a is the phase-A voltage, U b is the phase-B voltage, U c is the phase-C voltage.

3. The method for suppressing the common-mode voltage of an NPC-type converter with active neutral-point potential control according to claim 2, wherein In step 2), the reference voltage vector V ref The synthesis formula is: Among them, 4. The method for suppressing the common-mode voltage of an NPC type converter with active neutral point potential control according to claim 1, characterized in that, In Step 3), the space vector expression V of the three-level NPC converter is: Among them, S a , S b and S c are the switching states of phase A, phase B and phase C of the three-level NPC converter respectively. S a , S b , S c ∈ (-1, 0, 1), and the switching states P, O, N are represented by 1, 0, -1 respectively; According to the magnitude of the basic space vector, it is divided into basic zero vectors, basic small vectors, basic medium vectors and basic large vectors. Among them, there are 3 basic zero vectors, 12 basic small vectors, 6 basic medium vectors and 6 basic large vectors.

5. The method for suppressing the common-mode voltage of an NPC-type converter with active neutral-point potential control according to claim 1, wherein In Step 5), first perform the synthesis of the virtual vector in the first large sector, and transfer the reference voltage vector to the first large sector of the space vector diagram of the three-level NPC converter through the rotation transformation method for other sectors; The common-mode voltage V generated by 27 space vectors com The formula is as follows: The formula for judging the position of the large sector is: Among them, the ceil() function represents rounding up. That is, the first large sector corresponds to θ∈(0~60°), the second large sector corresponds to θ∈(60°~120°), the third large sector corresponds to θ∈(120°~180°), the fourth large sector corresponds to θ∈(180°~240°), the fifth large sector corresponds to θ∈(240°~300°), and the sixth large sector corresponds to θ∈(300°~360°); The rotation transformation formula for projecting other sectors onto the first sector is: S = θ - 60°(N - 1) (7) where S represents the projected reference voltage vector V ref corresponding to the position of the first major sector; In addition, according to the following formula, analyze the influence of the basic voltage vector on the neutral point current: i np =(1 - |S a |)i a +(1 - |S b |)i b +(1 - |S c |)i c (8) Among them, i a , i b , i c are the three-phase currents of the converter; According to the analysis of Equation (8), it can be known that the midpoint current i generated by the zero vector and the large vector np is 0, which has no effect on the midpoint potential in principle. Only the midpoint current generated by the small vector and the medium vector needs to be considered; According to the synthesized virtual vector, each large sector is further divided into 5 small sectors. Taking the first large sector as an example, the small sectors are divided by the following formula: Among them, L1 - L5 are the regional division lines.

6. The method for suppressing the common-mode voltage of an NPC-type converter with active neutral-point potential control according to claim 1, wherein In step 6), at this time, the virtual middle vector V VM generates a midpoint current i np which is: i np = ki a + (1 - 2k)i b + ki b = (1 - 3k)i b (12) Thus, the charge Q flowing out of the capacitor neutral point under the action of the virtual middle vector is obtained as: When k = 1 / 3, the charge Q flowing out of the neutral point is 0, and at this time, the neutral point current generated by the virtual middle vector is 0; Set A as an allowable voltage deviation value, and let ΔV be the voltage difference between the upper and lower two capacitors on the DC side, that is, ΔV = U CP - U CN , divide the value range of k into two segments of (0~1 / 3) and (1 / 3~1 / 2) with 1 / 3 as the boundary, and set k1 ∈ (0~1 / 3), k2 ∈ (1 / 3~1 / 2); the specific adjustment principle of k is as follows: (1) When -A ≤ ΔV ≤ A, it indicates that the midpoint voltage fluctuation is within the allowable range, and there is no need to suppress the midpoint voltage fluctuation at this time; (2) When ΔV > A, it indicates that the midpoint voltage fluctuation value exceeds the allowable range. At this time, it is necessary to make U CP become smaller, U CN become larger. Then, it is necessary to make the outflowing midpoint charge Q < 0. It can be seen from Equation (13) that when i b < 0, let k = k1; when i b > 0, let k = k2, which can make the charge Q < 0; (3) When ΔV < -A, it indicates that the midpoint voltage fluctuation value exceeds the allowable range. At this time, it is necessary to increase U CP and decrease U CN . Then, it is necessary to make the midpoint outflow charge Q > 0; also, it can be seen from Equation (13) that when i b < 0, let k = k2; when i b > 0, let k = k1, which can make the charge Q > 0; Therefore, by adjusting the distribution coefficient k, the situation where the midpoint voltage has an offset can be controlled originally.

7. The method for suppressing the common-mode voltage of an NPC-type converter with active neutral-point potential control according to claim 1, wherein In step 7), taking the third smallest sector of the first largest sector as an example, the virtual vector action time T is calculated by formula (14). VS1 , T VL1 , T VM ; where T S is the acting time of the reference voltage vector V ref ; To simplify the calculation, the g-h coordinate system is selected to calculate the action time of each virtual vector. The formula for converting the α-β coordinate system to the g-h coordinate system is: Taking the third smallest sector in the first large sector as an example, Equation (14) is expressed in the g-h coordinate system as: Among them, V VM(g) , V VM(h) are respectively the components of the medium vector on the V VM on the g and h axes; The action time of the virtual vector is calculated from Equation (16): Similarly, the action time of the virtual vector when the reference voltage vector V ref is located in other small sectors of the first large sector can be calculated. The action time of the virtual vectors in other large sectors can be obtained in the same way by rotation transformation to the first large sector.

8. The method for suppressing the common-mode voltage of the NPC type converter with active neutral point potential control according to claim 1, wherein In step 8), the principle for determining the switching sequence of each phase switch tube is that there should be a smooth switching between the vectors in each sector, that is, the switching state needs to follow the conversion between P and O or N and O, and the switching frequency is the lowest.

Images