Common-mode voltage suppression method for three-level inverter based on equivalent zero vector regulation

Abstract

The present invention discloses a method for suppressing common-mode voltage in a three-level inverter based on equivalent zero vector regulation. The method comprises the following steps: 1. establishing a mathematical model for a three-level NPC energy storage inverter to determine the influencing mechanism of common-mode voltage; 2. selecting a neutral vector and a redundant zero vector as basic synthetic vectors to minimize the common-mode voltage generated during the switching process. The 60° region between each pair of adjacent neutral vectors is divided into a sector. A pair of adjacent neutral vectors of equal amplitude and opposite phase in each sector is used as an equivalent zero vector to mitigate the common-mode voltage jump when the zero vector is applied; 3. deriving a time allocation factor k through a neutral current PI controller to adjust the application time of the equivalent zero vector and the redundant zero vector to reduce the average neutral current. By optimizing the selection of vectors and the application time of the equivalent zero vector and the redundant zero vector, the present invention suppresses the common-mode voltage and neutral point potential offset under full modulation, thus helping to maintain the stable operation of the energy storage system.

Description

Technical Field

[0001] The invention relates to a three-level inverter common mode voltage suppression method based on equivalent zero vector regulation, and belongs to the field of energy storage inverter control. Background Art

[0002] Compared to two-level inverters, three-level inverters have higher withstand voltage ratings and lower output voltage harmonics. They are widely used in energy storage systems, photovoltaic and wind power generation, and variable-frequency motor drive. Because the inverter's output characteristics at any given moment depend on the switching state, each switching state corresponds to a specific common-mode voltage. The high-frequency switching of the switching states at the switching frequency causes the inverter to output a high-frequency common-mode voltage. This common-mode voltage generates electromagnetic interference (EMI) on surrounding electrical and communications equipment, leading to a series of electromagnetic compatibility (EMC) issues. Furthermore, this high-frequency common-mode voltage acting on the parasitic capacitance between the inverter and ground can easily generate common-mode leakage current. This high-frequency common-mode leakage current not only generates grid-connected current harmonics, degrading power quality, but also easily causes electromagnetic interference and other issues, potentially threatening personal safety.

[0003] Common-mode voltage can be reduced through software or hardware. Hardware methods can use passive or active devices to suppress the inverter's common-mode voltage, but this approach increases the size of the inverter system. Software methods primarily suppress common-mode voltage by improving the inverter's modulation strategy. For example, a vector with zero common-mode voltage can be selected for modulation to achieve common-mode voltage suppression. However, this can cause a DC midpoint potential offset, potentially leading to overvoltage on the upper or lower DC capacitors. Alternatively, the SVPWM modulation strategy can be improved, using a three-level midpoint voltage balance control strategy that combines large, medium, and small vectors. This control method overcomes midpoint voltage offsets caused by non-redundant small vectors in low-modulation regimes and fully considers the impact of the medium vector on midpoint voltage offsets in high-modulation regimes, achieving refined control of midpoint voltage balance. However, this method is computationally complex and unsuitable for online control. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, the present invention proposes a common-mode voltage suppression method for a three-level inverter based on equivalent zero vector regulation, in order to achieve unified control of the common-mode voltage under full modulation, thereby avoiding the complexity of using different vector modulation methods for low and high modulation, effectively reducing electromagnetic interference to surrounding equipment, and improving the safe operation performance of the inverter.

[0005] In order to achieve the above-mentioned object, the present invention adopts the following technical solutions:

[0006] The present invention provides a method for suppressing common-mode voltage of a three-level inverter based on equivalent zero vector regulation, which comprises the following steps:

[0007] Step 1: Based on the topology of the three-level NPC energy storage inverter, establish the expression of the three-phase bridge arm output voltage, and obtain the relationship between the common mode voltage, the midpoint potential and the switching function;

[0008] Step 2: Analyze the relationship between the amplitude of different types of basic composite vectors and the common-mode voltage. With the goal of minimizing the common-mode voltage generated during the switching process, select six neutral vectors and one redundant zero vector from the different types of basic composite vectors as the basic composite vectors.

[0009] Step 3: Divide the 60° area between each two adjacent mid vectors into a sector, so that the α-β coordinate system is divided into six sectors, and the reference voltage vector V ref The angle θ with the positive semi-axis of α determines the reference voltage vector V ref In the sector where the voltage is located, four more mid-point vectors are selected from the basic synthetic vector and synthesized together with a redundant zero vector to form the corresponding reference voltage vector V ref , among the four mid-vectors, the two mid-vectors on the sector boundary are used as the two main vectors, and a pair of mid-vectors with equal amplitude and opposite phase adjacent to the sector are used as the two equivalent zero vectors;

[0010] Step 4: By establishing a relationship between the neutral current, the switching function, and the three-phase current value, the neutral current corresponding to the selected basic synthetic vector is obtained;

[0011] Step 5: Establish the average neutral current I o The expression of , and the time distribution factor k is obtained by using the neutral current PI controller, so that the sum of the equivalent zero vector action time is kT ez0 , the redundant zero vector action time is (1-k)T ez0 ; Among them, T ez0 It represents the sum of the action time of the equivalent zero vector and the redundant zero vector;

[0012] Step 6: Based on the parallelogram rule, establish the relationship between the basic synthetic vector of each sector and its action time, and calculate the duty cycle of the O state of each phase in different sectors and the action time of the selected basic synthetic vector.

[0013] The method for suppressing common-mode voltage of a three-level inverter based on equivalent zero vector regulation according to the present invention is also characterized in that step 1 comprises:

[0014] Using formula (1), we can get the expression of the three-phase bridge arm output voltage:

[0015]

[0016] In formula (1), U AO ,U BO ,U COis the output voltage of the three-phase bridge arm, ΔU C is the offset value of the midpoint potential, U dc is the voltage value on the DC side; S A , S B , S C They represent the switching functions of the A, B, and C phase bridge arms respectively, and we have:

[0017]

[0018] In formula (2): S y Represents the y-phase switching function, P represents the y-phase switching function in the first upper bridge arm switch tube Q y1 , the second upper arm switch tube Q y2 When the switch is on, O represents the y-phase switching function in the second upper arm switch tube Q y2 、The first lower bridge arm switch tube Q y3 The state when it is turned on, N represents the first lower bridge arm switch tube Q of the y-phase switching function y3 , the second lower bridge arm switch tube Q y4 When conducting, y = A, B, C;

[0019] Using formula (3) to get the common mode voltage U com The relationship between the midpoint potential and the switching function:

[0020]

[0021] In step 2, six neutral vectors and one redundant zero vector are selected as basic composite vectors, including:

[0022] The switching state of the three-phase switching function is OOO corresponding to the redundant zero vector V0;

[0023] The switching state of the three-phase switching function is the center vector V1 corresponding to PON;

[0024] The switching state of the three-phase switching function is the center vector V2 corresponding to OPN;

[0025] The switching state of the three-phase switching function is the middle vector V3 corresponding to NPO;

[0026] The switching state of the three-phase switching function is the middle vector V4 corresponding to NOP;

[0027] The switching state of the three-phase switching function is the middle vector V5 corresponding to ONP;

[0028] The switching state of the three-phase switching function is the center vector V6 corresponding to PNO.

[0029] The step 3 comprises:

[0030] When θ∈[-π / 6,π / 6], the reference voltage vector V ref Located in the first sector, two main vectors are selected as V1 and V6, two equivalent zero vectors are selected as V2 and V5, and one redundant zero vector is selected as V0 to synthesize the reference voltage vector V ref ;

[0031] When θ∈[π / 6,π / 2], the reference voltage vector V ref Located in the second sector, two main vectors are selected as V2 and V1, two equivalent zero vectors are selected as V3 and V6, and one redundant zero vector is selected as V0 to synthesize the reference voltage vector V ref ;

[0032] When θ∈[π / 2,π5 / 6], the reference voltage vector V ref Located in the third sector, two main vectors are selected as V3 and V2, two equivalent zero vectors are selected as V1 and V4, and one redundant zero vector is selected as V0 to synthesize the reference voltage vector V ref ;

[0033] When θ∈[π5 / 6,π7 / 6], the reference voltage vector V ref Located in the fourth sector, two main vectors are selected as V4 and V3, two equivalent zero vectors are V5 and V2, and one redundant zero vector is V0 to synthesize the reference voltage vector V ref ;

[0034] When θ∈[π7 / 6,π3 / 2], the reference voltage vector V ref Located in the fifth sector, two main vectors are selected as V5 and V4, two equivalent zero vectors are selected as V6 and V3, and one redundant zero vector is selected as V0 to synthesize the reference voltage vector V ref ;

[0035] When θ∈[π3 / 2,π11 / 6], the reference voltage vector V ref Located in the sixth sector, two main vectors are selected as V6 and V5, two equivalent zero vectors are selected as V4 and V1, and one redundant zero vector is selected as V0 to synthesize the reference voltage vector V ref .

[0036] In step 5, the neutral current expression is obtained using equation (4):

[0037] i o =(1-|S A |)i A +(1-|S B |)i B +(1-|S C |)i C (4)

[0038] In formula (4), iA 、i B 、i C is the three-phase bridge arm current value, i o is the neutral current value;

[0039] According to the selected basic synthetic vector, the corresponding neutral current is obtained using formula (4):

[0040] The neutral current i corresponding to the redundant zero vector V0 of the three-phase switch state OOO o =0;

[0041] The neutral current i corresponding to the neutral vector V1 of the three-phase switch state PON o =i B ;

[0042] The neutral current i corresponding to the neutral vector V2 of the three-phase switch state OPN o =i A ;

[0043] The neutral current i corresponding to the neutral vector V3 of the three-phase switch state NPO o =i C ;

[0044] The neutral current i corresponding to the neutral vector of the three-phase switch state NOP o =i B ;

[0045] The neutral line current i corresponding to the neutral vector V5 of the three-phase switch state ONP o =i A ;

[0046] The neutral current i corresponding to the neutral vector V6 of the three-phase switch state PNO o =i C .

[0047] In step 5, a switching period T is obtained by using formula (5) s Average neutral current I o The expression:

[0048] I o =D Ao i A +D Bo i B +D Co i C (5)

[0049] In formula (5): i A 、i B 、i C are the output current values ​​of the A, B, and C phase bridge arms respectively, and D Ao 、DBo 、D Co are the duty cycles of the A, B, and C phase switching functions in the O state, respectively;

[0050] The expression of the neutral current PI controller is established using equation (6):

[0051]

[0052] In formula (6), is the given value of the average neutral current, k p is the proportionality coefficient, k i is the integral coefficient, s is a variable in the complex frequency domain, and represents the time derivative.

[0053] The step 6 comprises:

[0054] When the reference voltage vector V ref When located in the first sector, the relationship between the selected basic synthetic vector and its action time is established using formula (7):

[0055]

[0056] In formula (7), |V ref | is the reference voltage vector V ref The amplitude, e jθ =cosθ+jsinθ is Euler's formula, j represents the imaginary part, T s is the switching period, T a 、T b is the action time of the two main vectors V1 and V6, T ez1 、T ez2 is the action time of two equivalent zero vectors V2 and V5, T0 is the action time of redundant zero vector V0, T ez0 is the sum of the action time of the equivalent zero vector and the redundant zero vector;

[0057] Using formula (8), we can get the basic synthetic vector selected by the first sector in one switching cycle T s The duty cycle of the O state of phases A, B, and C is:

[0058]

[0059] The action time of the basic synthetic vector selected by the first sector is obtained using formula (9):

[0060]

[0061] In formula (9), m represents the modulation index, and m=2|V ref | / U dc ;

[0062] When the reference voltage vector V ref When located in the second sector, the relationship between the selected basic synthetic vector and its action time is established using formula (10):

[0063]

[0064] In formula (10), T a 、T b is the action time of the two main vectors V2 and V1, T ez1 、T ez2 is the action time of the two equivalent zero vectors V3 and V6, T0 is the action time of the redundant zero vector V0, T ez0 is the sum of the action time of the equivalent zero vector and the redundant zero vector;

[0065] Using formula (11), we can get the basic synthetic vector selected by the second sector in one switching cycle T s The duty cycle of the O state of phases A, B, and C is:

[0066]

[0067] The action time of the basic synthetic vector selected by the second sector is obtained using formula (12):

[0068]

[0069] When the reference voltage vector V ref When located in the third sector, the relationship between the selected basic synthetic vector and its action time is established using formula (13):

[0070]

[0071] In formula (13), T a 、T b is the action time of the two main vectors V3 and V2, T ez1 、T ez2 is the action time of the two equivalent zero vectors V4 and V1, T0 is the action time of the redundant zero vector V0, T ez0 is the sum of the action time of the equivalent zero vector and the redundant zero vector;

[0072] Using formula (14), we can get the basic synthetic vector selected by the third sector in one switching cycle T s The duty cycle of the O state of phases A, B, and C is:

[0073]

[0074] The action time of the basic synthetic vector selected by the third sector is obtained using formula (15):

[0075]

[0076] When the reference voltage vector V ref When located in the fourth sector, the relationship between the selected basic synthetic vector and its action time is established using formula (16):

[0077]

[0078] In formula (16), T a 、T b is the action time of the two main vectors V4 and V3, T ez1 、T ez2 is the action time of two equivalent zero vectors V5 and V2, T0 is the action time of redundant zero vector V0, T ez0 is the sum of the action time of the equivalent zero vector and the redundant zero vector;

[0079] Using formula (17), we can get the basic synthetic vector selected by the fourth sector in one switching cycle T s The duty cycle of the O state of phases A, B, and C is:

[0080]

[0081] The action time of the basic synthetic vector selected in the fourth sector is obtained using formula (18):

[0082]

[0083] When the reference voltage vector V ref When located in the fifth sector, the relationship between the selected basic synthetic vector and its action time is established using formula (19):

[0084]

[0085] In formula (19), T a 、T b is the action time of the two main vectors V4 and V5, T ez1 、T ez2 is the action time of the two equivalent zero vectors V6 and V3, T0 is the action time of the redundant zero vector V0, T ez0 is the sum of the action time of the equivalent zero vector and the redundant zero vector;

[0086] Using formula (20), we can get the basic synthetic vector selected by the fifth sector in one switching cycle T s The duty cycle of the O state of phases A, B, and C is:

[0087]

[0088] The action time of the basic synthetic vector selected in the fifth sector is obtained using formula (21):

[0089]

[0090] When the reference voltage vector V ref When located in the sixth sector, the relationship between the selected basic synthetic vector and its action time is established using formula (22):

[0091]

[0092] In formula (22), T a 、T b is the action time of the two main vectors V6 and V5, T ez1 、T ez2 is the action time of two equivalent zero vectors V1 and V4, T0 is the action time of redundant zero vector V0, T ez0 is the sum of the action time of the equivalent zero vector and the redundant zero vector;

[0093] Using formula (23), we can get the basic synthetic vector selected by the sixth sector in one switching cycle T s The duty cycle of the O state of phases A, B, and C is:

[0094]

[0095] The action time of the basic synthetic vector selected in the sixth sector is obtained using formula (24):

[0096]

[0097] The electronic device of the present invention includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the three-level inverter common-mode voltage suppression method, and the processor is configured to execute the program stored in the memory.

[0098] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program executes the steps of the three-level inverter common-mode voltage suppression method when the computer program is executed by a processor.

[0099] Compared with the prior art, the present invention has the following beneficial effects:

[0100] 1. The present invention uses only six neutral vectors and one redundant zero vector to avoid causing high common-mode voltage as basic synthetic vectors, and selects a pair of neutral vectors with equal amplitude and opposite phase to synthesize an equivalent zero vector, thereby slowing down the sudden change of common-mode voltage when the zero vector acts.

[0101] 2. The present invention adjusts the action time of the equivalent zero vector and the zero vector through the time allocation factor k, thereby effectively reducing the average neutral current and the common-mode voltage caused by the midpoint potential offset. Compared with the traditional SVPWM modulation strategy, it optimizes the vector selection and action time, and improves the balancing ability of the midpoint potential.

[0102] 3. The present invention abandons the complex large-sector / small-sector division method and divides the 60° area between each two adjacent center vectors into one sector. Based on the different sectors where the target reference voltage vector is located, four center vectors and one redundant zero vector are specifically selected for synthesis. The two center vectors at the sector boundary serve as the main vectors, and an equivalent zero vector is synthesized with a pair of center vectors of equal amplitude and opposite phase adjacent to the sector, thereby greatly reducing the complexity of vector synthesis. BRIEF DESCRIPTION OF THE DRAWINGS

[0103] Figure 1 This is a schematic diagram of the circuit structure of a three-level NPC type energy storage inverter according to the present invention;

[0104] Figure 2 A sector diagram of the basic composite vector of the present invention;

[0105] Figure 3 A first sector vector composite diagram of the present invention;

[0106] Figure 4 A simplified diagram of the bridge arm switches of a three-level NPC inverter according to the present invention;

[0107] Figure 5 This is the structure diagram of the PI controller for the neutral current of the first sector in the present invention.

[0108] Figure 6 This is a three-phase switch state sequence diagram of the first sector of the present invention;

[0109] Figure 7 This is a comparison diagram of the midpoint potential fluctuations between the 4MV modulation strategy and the modulation strategy of the present invention;

[0110] Figure 8 This is a comparison diagram of the common mode voltage between the 4MV modulation strategy and the modulation strategy of the present invention. DETAILED DESCRIPTION

[0111] In this embodiment, to address the issues of high common-mode voltage and large midpoint potential offset in the output bridge arms of three-level inverters caused by traditional vector modulation strategies, a common-mode voltage suppression method for three-level inverters based on equivalent zero vector regulation is proposed. First, a mathematical model of a three-level NPC inverter is established to derive the relationship between the common-mode voltage, the switching function, and the DC side midpoint potential offset. Then, based on the principles of avoiding high common-mode voltage and reducing vector synthesis complexity, six neutral vectors and one redundant zero vector are selected as basic synthesis vectors. The 60° region between each two adjacent neutral vectors is divided into a sector. A pair of neutral vectors of equal amplitude and opposite phase adjacent to the sector are used to synthesize an equivalent zero vector to mitigate sudden changes in common-mode voltage when the zero vector is applied. Finally, the equivalent zero vector and the zero vector application time are adjusted using a time allocation factor, k, to reduce the average neutral current flowing into and out of the neutral point, thereby reducing the midpoint potential offset and effectively suppressing the amplitude of the common-mode voltage output by the three-level inverter.

[0112] The circuit model of the three-level NPC inverter is as follows: Figure 1 As shown, first give the definition of common mode voltage as formula (1):

[0113]

[0114] In formula (1), U com is the common mode voltage, U AO ,U BO ,U CO is the three-phase bridge arm output voltage.

[0115] According to Kirchhoff's voltage law, the circuit voltage equation of the three-level NPC inverter can be obtained as shown in formula (2):

[0116]

[0117] In formula (2), i y is the output current of the y-phase bridge arm of the three-level inverter, y=A, B, C; i x The x-phase load-side output current of the three-level inverter, x∈{a,b,c}; L f is the filter inductor, R is the load resistance; U NO It is the voltage from the neutral point of the load side voltage to the midpoint of the DC side.

[0118] The three-phase bridge arm switching function is as shown in formula (3):

[0119]

[0120] In formula (3): S y Represents the y-phase switching function, P represents the y-phase switching function in the first upper bridge arm switch tube Q y1 , the second upper arm switch tube Qy2 When the switch is on, O represents the y-phase switching function in the second upper arm switch tube Q y2 、The first lower bridge arm switch tube Q y3 The state when it is turned on, N represents the first lower bridge arm switch tube Q of the y-phase switching function y3 , the second lower bridge arm switch tube Q y4 When conducting, y = A, B, C;

[0121] Among them, the output voltage of the three-phase bridge arm can be expressed by the switching function as shown in formula (4):

[0122]

[0123] In formula (4), ΔU C is the midpoint potential offset value, U dc is the DC side voltage value;

[0124] Substituting formula (4) into (1) yields formula (5):

[0125]

[0126] According to the expression of formula (5), the common-mode voltage of the three-level inverter is determined by the basic synthetic vector corresponding to different types of switching functions and the midpoint potential offset value. Therefore, to suppress the common-mode voltage, it is necessary not only to avoid using the basic synthetic vector that causes high common-mode voltage, but also to suppress the midpoint potential fluctuation.

[0127] Among them, the relationship between the switching state of the three-level inverter, the basic synthetic vector and the common mode voltage is shown in Table 1:

[0128] Table 1 Relationship between different switch states, basic synthesis vectors and common mode voltage

[0129]

[0130] As shown in Table 1, large vectors, small vectors, and zero vectors (PPP, NNN) all generate high common-mode voltages. Therefore, the present invention avoids using basic synthetic vectors that cause high common-mode voltages and selects six medium vectors and one redundant zero vector as basic synthetic vectors, including:

[0131] The switching state of the three-phase switching function is OOO corresponding to the redundant zero vector V0;

[0132] The switching state of the three-phase switching function is the center vector V1 corresponding to PON;

[0133] The switching state of the three-phase switching function is the center vector V2 corresponding to OPN;

[0134] The switching state of the three-phase switching function is the middle vector V3 corresponding to NPO;

[0135] The switching state of the three-phase switching function is the middle vector V4 corresponding to NOP;

[0136] The switching state of the three-phase switching function is the middle vector V5 corresponding to ONP;

[0137] The switching state of the three-phase switching function is the middle vector V6 corresponding to PNO;

[0138] The 60° area between each two adjacent mid-vectors is divided into a sector, so that the α-β coordinate system is divided into six sectors, and according to the reference voltage vector V ref The angle θ with the positive semi-axis of α determines the reference voltage vector V ref The sector where it is located is divided into the following diagrams: Figure 2 .

[0139] The present invention is based on the reference voltage vector V ref In different sectors, four mid-vectors and one redundant zero vector are selected for synthesis. Among the four mid-vectors, the two mid-vectors on the sector boundary are used as the two main vectors, and a pair of mid-vectors with equal amplitude and opposite phases adjacent to the sector are used as two equivalent zero vectors to slow down the jump of the common-mode voltage when the zero vector acts. When the reference voltage vector V ref The basic synthetic vectors selected when located in different sectors are as follows:

[0140] When θ∈[-π / 6,π / 6], the reference voltage vector V ref Located in the first sector, two main vectors are selected as V1 and V6, two equivalent zero vectors are selected as V2 and V5, and one redundant zero vector is selected as V0 to synthesize the reference voltage vector V ref ;

[0141] When θ∈[π / 6,π / 2], the reference voltage vector V ref Located in the second sector, two main vectors are selected as V2 and V1, two equivalent zero vectors are selected as V3 and V6, and one redundant zero vector is selected as V0 to synthesize the reference voltage vector V ref ;

[0142] When θ∈[π / 2,π5 / 6], the reference voltage vector V ref Located in the third sector, two main vectors are selected as V3 and V2, two equivalent zero vectors are selected as V1 and V4, and one redundant zero vector is selected as V0 to synthesize the reference voltage vector V ref ;

[0143] When θ∈[π5 / 6,π7 / 6], the reference voltage vector V ref Located in the fourth sector, two main vectors are selected as V4 and V3, two equivalent zero vectors are V5 and V2, and one redundant zero vector is V0 to synthesize the reference voltage vector Vref ;

[0144] When θ∈[π7 / 6,π3 / 2], the reference voltage vector V ref Located in the fifth sector, two main vectors are selected as V5 and V4, two equivalent zero vectors are selected as V6 and V3, and one redundant zero vector is selected as V0 to synthesize the reference voltage vector V ref ;

[0145] When θ∈[π3 / 2,π11 / 6], the reference voltage vector V ref Located in the sixth sector, two main vectors are selected as V6 and V5, two equivalent zero vectors are selected as V4 and V1, and one redundant zero vector is selected as V0 to synthesize the reference voltage vector V ref ;

[0146] like Figure 3 The first sector vector composite diagram is shown. When V ref When in the first sector, V ref The rotation angle θ∈[-π / 6,π / 6], the vectors involved in the synthesis are V0, V1, V2, V5, V6, where the main vectors are V1 and V6, V2 and V5 are used as equivalent zero vectors, and the relationship established by the parallelogram law is shown in formula (6):

[0147]

[0148] In formula (6), |V ref | is the reference voltage vector V ref The amplitude, e jθ =cosθ+jsinθ is Euler's formula, j represents the imaginary part, T s is the switching period, T a 、T b is the action time of the two main vectors V1 and V6, T ez1 、T ez2 is the action time of two equivalent zero vectors V2 and V5, T0 is the action time of redundant zero vector V0, T ez0 is the sum of the action time of the equivalent zero vector and the redundant zero vector;

[0149] According to formula (7), the expressions of DC side capacitor voltage, capacitor current value and neutral line current are established:

[0150]

[0151] In formula (7), it is assumed that the two capacitors on the DC side are exactly the same, that is, C1=C2=C, U C1 is the voltage of capacitor C1, U C2 is the voltage of capacitor C2, i C1 is the current flowing through capacitor C1, iC2 is the current flowing through capacitor C2, i o is the value of the neutral current;

[0152] like Figure 4 This is a simplified diagram of the three-level NPC inverter switch. When the three-level inverter is working, only the O state will generate a neutral current on the neutral line. According to formula (8), the expression of the neutral current, three-phase current value, and three-phase switch function is established:

[0153] i o =(1-|S A |)i A +(1-|S B |)i B +(1-|S C |)i C (8)

[0154] In formula (8), i A 、i B 、i C are the current values ​​output by the A, B, and C phase bridge arms respectively;

[0155] From the analysis of equations (7) and (8), it can be seen that the fluctuation of the midpoint potential is mainly caused by the imbalance of the neutral current in charging and discharging the DC side capacitor. Therefore, the study of suppressing the fluctuation of the midpoint potential can be turned to the study of reducing the neutral current. The six midpoint vectors V m The corresponding neutral current is shown in Table 2, m = 1, 2, 3, 4, 5, 6;

[0156] Table 2 Relationship between vector and neutral current

[0157]

[0158] Since the three-phase load is symmetrical, the neutral current corresponding to the redundant zero vector (OOO) is 0, as shown in formula (9):

[0159] i o =i A +i B +i C =0 (9)

[0160] The average neutral current in one switching cycle is obtained using formula (10):

[0161] I o =D Ao i A +D Bo i B +D Co i C (10)

[0162] In formula (10): I o is the average value of the neutral current in one switching cycle, i A 、i B 、i C are the output current values ​​of the A, B, and C phase bridge arms respectively, and D Ao 、D Bo 、D Co are the duty cycles of the A, B, and C phase switching functions in the O state, respectively;

[0163] From the above analysis, we can see that in the first sector, the switching state of the middle vector corresponding to the main vector V1 is PON, and the generated neutral current is i B , the switching state of the middle vector corresponding to the main vector V6 is PNO, and the generated neutral current is i C , the switching state of the neutral vector corresponding to the equivalent zero vector is OPN / ONP, and the neutral current generated is i A , the switch state corresponding to the redundant zero vector V0 is OOO, and the resulting neutral current is 0. If I o = 0, it can be considered that the DC side midpoint potential does not fluctuate or fluctuates very little within a switching cycle. Therefore, in order to suppress the midpoint potential offset caused by the neutral current, such as Figure 5 As shown, in the first sector, through the adjustment of the neutral current PI controller, the time allocation factor k is obtained as shown in formula (11):

[0164]

[0165] In formula (11): k is the time allocation factor, is the target value of the average neutral current, k p is the proportionality coefficient, k i is the integral coefficient, s is a variable in the complex frequency domain, representing the time derivative;

[0166] In the first sector, the time allocation factor k obtained by formula (11) is used to balance the action time of the equivalent zero vector and the redundant zero vector, so that the sum of the action time of the equivalent zero vector is kT ez0 , the action time of redundant zero vector is (1-k)T ez0 , thereby effectively reducing the average neutral current and lowering the common-mode voltage caused by the midpoint potential offset.

[0167] Combining equations (7) to (11), we can obtain the basic synthetic vector of the first sector in one switching cycle T s The duty ratio of the O state of phases A, B, and C is shown in formula (12):

[0168]

[0169] From the above analysis, we can see that in the first sector, the action time of the basic synthetic vector selected as shown in formula (13) is obtained:

[0170]

[0171] In formula (13), the modulation index m=2|V ref | / U dc .

[0172] The present invention adopts a nine-segment switching sequence. The action order of the vector in the first sector is first counterclockwise rotation upward and then clockwise rotation downward, that is, starting from the equivalent zero vector V5, passing through the main vectors V6 and V1 counterclockwise, then reaching another equivalent zero vector V2, then reaching the redundant zero vector V0, and finally starting from V2 and returning to the starting equivalent zero vector V5. The vector switching order when the reference voltage vector is located in the first sector is V5→V6→V1→V2→V0→V2→V1→V6→V5. From this, the switching state sequence diagram of the three phases of the first sector can be obtained, as shown in FIG. Figure 6 shown.

[0173] The reference voltage vectors are located in different sectors, and the vector switching order is as follows:

[0174] V ref The vector switching order in the first sector is: V5→V6→V1→V2→V0→V2→V1→V6→V5;

[0175] V ref The vector switching order in the second sector is: V6 → V1 → V2 → V3 → V0 → V3 → V2 → V1 → V6;

[0176] V ref The vector switching order in the third sector is: V1→V2→V3→V4→V0→V4→V3→V2→V1;

[0177] V ref The vector switching order in the fourth sector is: V2→V3→V4→V5→V0→V5→V4→V3→V2;

[0178] V ref The vector switching order in the fifth sector is: V3→V4→V5→V6→V0→V6→V5→V4→V3;

[0179] V ref The vector switching order in the sixth sector is: V4 → V5 → V6 → V1 → V0 → V1 → V6 → V5 → V4;

[0180] Similarly, when the reference voltage vector is located in other sectors, the action time of the selected basic synthetic vector and the time in a switching cycle T can be obtained.s The duty cycle of the O state of phases A, B, and C is shown in Table 3.

[0181] Table 3 The action time of the basic synthetic vector in different sectors of the present invention and the duty cycle of the O state of each phase switch function

[0182]

[0183] The three-phase bridge arms of the three-level NPC inverter perform vector synthesis according to the vector switching sequence of different sectors and the action time of each basic synthetic vector and the duty cycle of each phase O state set in Table 3, thereby achieving the suppression of the midpoint potential offset and common-mode voltage of the three-level inverter.

[0184] The following are the main parameters in this example:

[0185]

[0186] In this embodiment, an electronic device includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the above method, and the processor is configured to execute the program stored in the memory.

[0187] In this embodiment, a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are executed.

[0188] In order to illustrate the suppression effect of the common mode voltage and DC side midpoint potential deviation of the solution proposed in the present invention, it is assumed that the modulation index m = 0.5 and the average neutral current The midpoint potential offset and common-mode voltage suppression effects of the vector modulation strategy in 4MV and the modulation method of the present invention were compared through Matlab / Simulink simulation.

[0189] Figure 7 The figure is a comparison of the midpoint potential fluctuations using the 4MV modulation strategy and the modulation method of the present invention, where the horizontal axis is time t / s and the vertical axis is the midpoint potential offset voltage value, that is, the voltage difference between the upper and lower capacitors, in V. Figure 7 It can be obtained that the midpoint potential offset amplitude of the 4MV modulation strategy is about 63V, and the midpoint potential offset amplitude of the modulation strategy of the present invention is about 8V. Compared with the 4MV modulation strategy, the midpoint potential offset value of the method proposed in the present invention is reduced by about 87.31%, verifying that the effect of suppressing the midpoint potential offset of the present invention is better.

[0190] Figure 8 The following is a comparison chart of the common mode voltage of the 4MV modulation strategy and the common mode voltage of the modulation method of the present invention, wherein the horizontal axis is time t / s and the vertical axis is the common mode voltage value in V. Figure 8It can be obtained that the common-mode voltage amplitude of the modulation strategy of the present invention is about 1.8V, and the common-mode voltage of the 4MV modulation strategy is about 25.2V. Compared with the 4MV modulation strategy, the midpoint potential offset value of the method proposed in the present invention is reduced by about 92.86%, verifying that the present invention has better common-mode voltage suppression effect while suppressing the midpoint potential offset.

Claims

1. A method for suppressing common-mode voltage of a three-level inverter based on equivalent zero vector regulation, characterized in that: The following steps are involved: Step 1: Based on the topology of the three-level NPC energy storage inverter, establish the expression of the three-phase bridge arm output voltage, and obtain the relationship between the common mode voltage, the midpoint potential and the switching function; Using formula (1), we can get the expression of the three-phase bridge arm output voltage: (1) In formula (1), is the output voltage of the three-phase bridge arm, is the offset value of the midpoint potential, is the voltage value on the DC side; , , Respectively The switching function of the phase bridge arm is: (2) In formula (2): express Phase switching function, P represents Phase switching function is in the first upper arm switch tube , the second upper arm switch tube The state when conducting, O means Phase switching function is in the second upper bridge arm switch tube , the first lower bridge arm switch tube The state when conducting, N represents Phase switch function first lower bridge arm switch tube , the second lower bridge arm switch tube The state when conducting, ; The first upper arm switch tube The top switch tube of the upper bridge arm and the second lower bridge arm switch tube It is the lowest switch tube of the lower bridge arm; Using formula (3) to get the common mode voltage The relationship between the midpoint potential and the switching function: (3) Step 2: Analyze the relationship between the amplitude of different types of basic composite vectors and the common-mode voltage. With the goal of minimizing the common-mode voltage generated during the switching process, select six neutral vectors and one redundant zero vector from the different types of basic composite vectors as the basic composite vectors. Step 3: Divide the 60° area between each two adjacent mid vectors into a sector, so that The coordinate system is divided into six sectors and is based on the reference voltage vector The angle θ with the positive semi-axis of α determines the reference voltage vector In the sector where the vector is located, four more vectors are selected from the basic synthetic vector and synthesized together with a redundant zero vector to form the corresponding reference voltage vector , among the four mid-vectors, the two mid-vectors on the sector boundary are used as the two main vectors, and a pair of mid-vectors with equal amplitude and opposite phase adjacent to the sector are used as the two equivalent zero vectors; Step 4: By establishing a relationship between the neutral current, the switching function, and the three-phase current value, the neutral current corresponding to the selected basic synthetic vector is obtained; Step 5: Use formula (5) to establish a switching cycle Average neutral current within The expression of the neutral current PI controller is established using equation (6), and the time distribution factor k is obtained, so that the sum of the equivalent zero vector action time is obtained as , the redundant zero vector action time is ;in, It represents the sum of the action time of the equivalent zero vector and the redundant zero vector; (5) In formula (5): 、 、 They are The current value output by the phase bridge arm, 、 、 They are The duty cycle of the phase switching function in the O state; (6) In formula (6), is the given value of the average neutral current, is the proportionality coefficient, is the integral coefficient, s is a variable in the complex frequency domain, representing the time derivative; Step 6: Based on the parallelogram rule, establish the relationship between the basic synthetic vector of each sector and its action time, and calculate the duty cycle of the O state of each phase in different sectors and the action time of the selected basic synthetic vector.

2. The method for suppressing common-mode voltage of a three-level inverter based on equivalent zero vector regulation according to claim 1, characterized in that: In step 2, six neutral vectors and one redundant zero vector are selected as basic composite vectors, including: The switching state of the three-phase switching function is OOO corresponding to the redundant zero vector ; The switching state of the three-phase switching function is the mid-vector corresponding to PON ; The switching state of the three-phase switching function is the middle vector corresponding to OPN ; The switching state of the three-phase switching function is the middle vector corresponding to NPO ; The switching state of the three-phase switching function is the middle vector corresponding to NOP ; The switching state of the three-phase switching function is the middle vector corresponding to ONP ; The switching state of the three-phase switching function is the mid-vector corresponding to PNO .

3. The method for suppressing common-mode voltage of a three-level inverter based on equivalent zero vector regulation according to claim 2, characterized in that: The step 3 comprises: when , reference voltage vector Located in the first sector, two main vectors are selected as 、 , the two equivalent zero vectors are 、 , a redundant zero vector is , used to synthesize the reference voltage vector ; when , reference voltage vector Located in the second sector, two main vectors are selected as 、 , the two equivalent zero vectors are 、 , a redundant zero vector is , used to synthesize the reference voltage vector ; when , reference voltage vector Located in the third sector, two main vectors are selected as 、 , the two equivalent zero vectors are 、 , a redundant zero vector is , used to synthesize the reference voltage vector ; when , reference voltage vector Located in the fourth sector, two main vectors are selected as 、 , the two equivalent zero vectors are 、 , a redundant zero vector is , used to synthesize the reference voltage vector ; when , reference voltage vector Located in the fifth sector, select two main vectors as 、 , the two equivalent zero vectors are 、 , a redundant zero vector is , used to synthesize the reference voltage vector ; when , reference voltage vector Located in the sixth sector, two main vectors are selected as 、 , the two equivalent zero vectors are 、 , a redundant zero vector is , used to synthesize the reference voltage vector .

4. The method for suppressing common-mode voltage of a three-level inverter based on equivalent zero vector regulation according to claim 3, characterized in that: In step 4, the neutral current expression is obtained using formula (4): (4) In formula (4), 、 、 is the current value output by the three-phase bridge arm, is the neutral current value; According to the selected basic synthetic vector, the corresponding neutral current is obtained using formula (4): Redundant zero vector of three-phase switch state OOO Corresponding neutral current =0; The three-phase switch state is the middle vector of PON Corresponding neutral current = ; The three-phase switch state is OPN. Corresponding neutral current = ; The three-phase switch state is the middle vector of NPO Corresponding neutral current = ; The neutral current corresponding to the neutral vector V4 of the three-phase switch state NOP = ; The three-phase switch state is ONP. Corresponding neutral current = ; The three-phase switch state is the middle vector of PNO Corresponding neutral current = .

5. The method for suppressing common-mode voltage of a three-level inverter based on equivalent zero vector regulation according to claim 4, characterized in that: The step 6 comprises: When the reference voltage vector When located in the first sector, the relationship between the selected basic synthetic vector and its action time is established using formula (7): (7) In formula (7), is the reference voltage vector The amplitude of is Euler's formula, j represents the imaginary part, is the switching period, 、 Two main vectors 、 The action time, 、 are two equivalent zero vectors 、 The action time, Redundant zero vector The action time, is the sum of the action time of the equivalent zero vector and the redundant zero vector; Using formula (8), we can get the basic synthetic vector selected by the first sector in one switching cycle: middle The duty cycle of the O state of the phase: (8) The action time of the basic synthetic vector selected by the first sector is obtained using formula (9): (9) In formula (9), represents the modulation index, and ; When the reference voltage vector When located in the second sector, the relationship between the selected basic synthetic vector and its action time is established using formula (10): (10) In formula (10), 、 Two main vectors 、 The action time, 、 are two equivalent zero vectors 、 The action time, Redundant zero vector The action time, is the sum of the action time of the equivalent zero vector and the redundant zero vector; Using formula (11), we can get the basic synthetic vector selected by the second sector in one switching cycle: middle The duty cycle of the O state of the phase: (11) The action time of the basic synthetic vector selected by the second sector is obtained using formula (12): (12) When the reference voltage vector When located in the third sector, the relationship between the selected basic synthetic vector and its action time is established using formula (13): (13) In formula (13), 、 Two main vectors 、 The action time, 、 are two equivalent zero vectors 、 The action time, Redundant zero vector The action time, is the sum of the action time of the equivalent zero vector and the redundant zero vector; Using formula (14), we can get the basic synthetic vector selected by the third sector in one switching cycle: middle The duty cycle of the O state of the phase: (14) The action time of the basic synthetic vector selected by the third sector is obtained using formula (15): (15) When the reference voltage vector When located in the fourth sector, the relationship between the selected basic synthetic vector and its action time is established using formula (16): (16) In formula (16), 、 Two main vectors 、 The action time, 、 are two equivalent zero vectors 、 The action time, Redundant zero vector The action time, is the sum of the action time of the equivalent zero vector and the redundant zero vector; Using formula (17), we can get the basic synthetic vector selected by the fourth sector in one switching cycle: middle The duty cycle of the O state of the phase: (17) The action time of the basic synthetic vector selected in the fourth sector is obtained using formula (18): (18) When the reference voltage vector When located in the fifth sector, the relationship between the selected basic synthetic vector and its action time is established using formula (19): (19) In formula (19), 、 Two main vectors 、 The action time, 、 are two equivalent zero vectors 、 The action time, Redundant zero vector The action time, is the sum of the action time of the equivalent zero vector and the redundant zero vector; Using formula (20), we can get the basic synthetic vector selected by the fifth sector in one switching cycle: middle The duty cycle of the O state of the phase: (20) The action time of the basic synthetic vector selected in the fifth sector is obtained using formula (21): (21) When the reference voltage vector When located in the sixth sector, the relationship between the selected basic synthetic vector and its action time is established using formula (22): (22) In formula (22), 、 Two main vectors 、 The action time, 、 are two equivalent zero vectors 、 The action time, Redundant zero vector The action time, is the sum of the action time of the equivalent zero vector and the redundant zero vector; Using formula (23), the basic synthetic vector selected by the sixth sector is obtained in one switching cycle: middle The duty cycle of the O state of the phase: (23) The action time of the basic synthetic vector selected in the sixth sector is obtained using formula (24): (24)。 6. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store a program that supports the processor to execute the three-level inverter common-mode voltage suppression method according to any one of claims 1 to 5, and the processor is configured to execute the program stored in the memory.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for suppressing common-mode voltage of a three-level inverter according to any one of claims 1 to 5 are executed.

Images