Midpoint Potential Control Method for Regulating Common Mode Voltage in T-Type Three-Level Inverter

Abstract

The present invention discloses a midpoint potential control method for adjusting the common mode voltage in a T-type three-level inverter, comprising the following steps: S1, defining a virtual basic vector to avoid the participation of a positive vector; S2, according to the virtual basic vector, using a space vector pulse width modulation method, and adjusting the action time of the virtual basic vector according to a time correction factor, to balance control the midpoint potential of the common mode voltage in the T-type three-level inverter; the present invention adopts a space vector pulse width modulation method, and adjusts the action time of the virtual basic vector according to the time correction factor to achieve control of the midpoint potential. Thus, the coordinated suppression of the midpoint potential and the common mode voltage according to the correction factor is achieved.

Description

Technical Field

[0001] The invention relates to the field of power electronic control, and in particular to a midpoint potential control method for regulating a common mode voltage in a T-type three-level inverter. Background Art

[0002] Neutral-point-clamped (NPC) three-level inverters are mainly faced with the problem of unbalanced midpoint potential. The midpoint potential refers to the potential at the connection point of the two capacitors on the DC side. In order to ensure the safe and reliable operation of the three-level inverter, it is necessary to ensure that the two capacitors of the same capacitance on the DC side must be balanced in voltage division, otherwise it will lead to unbalanced midpoint potential, damage the life of the capacitors, and increase the voltage stress of the switching devices. In severe cases, the switching devices may be directly damaged, and the output waveform of the inverter will be distorted, the harmonic content will increase, and the load will be harmed. Secondly, in NPC inverters, the suppression of common mode voltage (CMV) is also an important issue. Higher CMV will cause a series of problems during the operation of the system. For example, higher CMV will induce a higher amplitude shaft voltage on the shaft of the motor, thereby forming a shaft current, damaging the insulation, and shortening the service life of the motor; high-frequency CMV will generate high-frequency leakage current, thereby causing serious electromagnetic interference to the surrounding electrical equipment that is working normally.

[0003] At present, the commonly used solutions are mainly divided into two categories: hardware improvement of topology and optimization of software algorithms. Compared with the hardware method of improving the topology, the optimization software algorithm has the advantages of convenient operation, simple experiment and low cost, so it is more favored by experts and scholars. The corresponding software algorithms mainly include: virtual space vector, reasonable adjustment of small vector action time and sequence, sliding mode control, zero-sequence component injection and other methods. Most of these research methods only focus on one aspect of midpoint potential balance and common mode voltage suppression, and there is little research on the coordinated control of the two. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the present invention provides a midpoint potential control method for adjusting the common mode voltage in a T-type three-level inverter, which solves the problem that only midpoint potential balance or common mode voltage suppression is considered in the existing T-type three-level inverter control.

[0005] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is: a midpoint potential control method for adjusting the common mode voltage in a T-type three-level inverter, comprising the following steps:

[0006] S1. Define a virtual basic vector to avoid the participation of positive vectors;

[0007] S2. According to the virtual basic vector, a space vector pulse width modulation method is adopted, and the action time of the virtual basic vector is adjusted according to the time correction factor to balance the midpoint potential of the common mode voltage in the T-type three-level inverter.

[0008] Furthermore, the definition of the virtual basic vector in step S1 includes:

[0009] V N0 =V OOO

[0010]

[0011]

[0012]

[0013] V NL1 =V PNN

[0014] V NL2 =V PPN

[0015] Among them, V N0 is the virtual basic zero vector, V NS1 is the first virtual basic vector, V NS2 is the second virtual basic vector, V NM1 is the virtual basic vector, V NL1 is the first virtual basic vector, V NL2 is the second virtual basic vector, V OOO is the original zero vector under OOO switching states, V OON is the original negative basic small vector in the OON switching state, V POO is the original negative basic small vector in POO switching states, V ONO is the original negative basic small vector in the OON switching state, V OPO is the original negative basic small vector in the OPO switching state, V PON is the original basic vector in the PON switching state, V OPN is the original basic center vector in the OPN switching state, V PNO is the original basic center vector in the PNO switching state, V PNN is the original basic vector in the PNN switching state, V PPN It is the original basic large vector under the PPN switching state.

[0016] The beneficial effects of the above further scheme are as follows: the virtual basic zero vector is the same as the original zero vector, the virtual basic large vector is the same as the original basic large vector, the virtual basic medium vector is synthesized by three original basic medium vectors, and the virtual basic small vector is synthesized by the original negative basic small vector and its two adjacent original negative basic small vectors, avoiding the participation of the positive vector, so the common mode voltage generated when the virtual basic small vector acts is U dc / 6, which is half of the original common-mode voltage.

[0017] Furthermore, the step S2 adjusts the action time of the virtual basic vector according to the time correction factor through the following formula:

[0018] T′2=fT2

[0019] Wherein, f is the time correction factor, T2 is the original action time of the vector in the virtual basic, and T′2 is the action time of the vector in the virtual basic after adjustment.

[0020] Furthermore, the time correction factor is obtained by solving the following formula:

[0021]

[0022] Where f is the time correction factor, C is the capacitance value, ΔU is the voltage difference between the two capacitors, that is, the midpoint potential, T2 is the original action time of the virtual basic vector, i b is the B phase load current.

[0023] The beneficial effect of the above further solution is that the virtual center vector of the present invention is synthesized by three adjacent center vectors, with the virtual center vector V NM1 For example, V NM1 The three basic voltage vectors V PON 、V OPN 、V PNO Synthesized, the midpoint currents generated by the three basic vectors are i b 、i a 、i c . Medium vector V PON With V PNO The sum of the midpoint currents generated is -i a , exactly the same as V OPN The generated midpoint current i a The modulus values ​​are equal but the directions are opposite, so the correction factor f is added to the action time of these three vectors, and the charge at the midpoint is controlled by adjusting the correction factor, thereby suppressing the fluctuation of the midpoint potential. Take the reference vector located in the fourth area in the first area as an example, because the virtual vector V NM1 and voltage vector V PON The direction is consistent, so the voltage vector V PON The time coefficient is (1+f), and the voltage vector VOPN 、V PNO The time coefficient is (1-f), so no matter how the correction factor f changes, it will not affect the direction and amplitude of the virtual vector.

[0024] In summary, the beneficial effects of the present invention are:

[0025] 1. The present invention defines a virtual basic vector to avoid the participation of a positive vector, so that the common mode voltage generated when the virtual basic small vector acts is U dc / 6, which is half of the original common-mode voltage, reduces the common-mode voltage.

[0026] 2. The space vector pulse width modulation method is adopted, and the action time of the virtual basic vector is adjusted according to the time correction factor to realize the control of the midpoint potential. Thus, the coordinated suppression of the midpoint potential and the common mode voltage is realized according to the correction factor. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is the topological structure diagram of T-type three-level inverter;

[0028] Figure 2 It is a space vector diagram;

[0029] Figure 3 A flow chart of a midpoint potential control method for adjusting the common mode voltage in a T-type three-level inverter;

[0030] Figure 4 It is a virtual space vector partition diagram;

[0031] Figure 5 It is the vector action time and output sequence diagram;

[0032] Figure 6 This is the comparison result diagram of midpoint potential simulation;

[0033] Figure 7 This is the comparison result diagram of common mode voltage simulation;

[0034] Figure 8 This is the waveform of the midpoint potential when the initial deviation is 10V. DETAILED DESCRIPTION

[0035] The specific implementation modes of the present invention are described below to facilitate those skilled in the art to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0036] The topology of the T-type three-level inverter is as follows: Figure 1It mainly includes 12 insulated gate bipolar transistors (IGBTs), each phase includes 4 switching devices IGBT, S m2 and S m3 Connect the midpoint and the load to form a bidirectional switch, S m1 and S m4 Connect the positive and negative poles of the DC bus voltage respectively, where m = a, b, c (one of the three phases). dc is the DC bus voltage, i a 、i b 、i c is the load three-phase current, C1 and C2 are DC side capacitors, i c1 and i c2 is the capacitor current. The T-type three-level inverter can output three levels of P, O, and N, so there are 3 three-phase 3 = 27 switch states, corresponding to 27 basic space vectors, such as Figure 2 According to the magnitude of the space vector, it is divided into four types: large vector, medium vector, small vector and zero vector.

[0037] like Figure 3 As shown, a midpoint potential control method for adjusting the common mode voltage in a T-type three-level inverter comprises the following steps:

[0038] S1. Define a virtual basic vector to avoid the participation of positive vectors;

[0039] S2. According to the virtual basic vector, a space vector pulse width modulation method is adopted, and the action time of the virtual basic vector is adjusted according to the time correction factor to balance the midpoint potential of the common mode voltage in the T-type three-level inverter.

[0040] When defining the virtual basic vector, the space vector diagram is divided into six large intervals every 60°, such as Figure 4 As shown, taking the first large interval as an example, the virtual vector is redefined as follows:

[0041] V N0 =V OOO

[0042]

[0043]

[0044]

[0045] V NL1 =V PNN

[0046] V NL2 =V PPN

[0047] Among them, V N0 is the virtual basic zero vector, V NS1 is the first virtual basic vector, V NS2 is the second virtual basic vector, V NM1 is the virtual basic vector, V NL1 is the first virtual basic vector, V NL2 is the second virtual basic vector, V OOO is the original zero vector under OOO switching states, V OON is the original negative basic small vector in the OON switching state, V POO is the original negative basic small vector in POO switching states, V ONO is the original negative basic small vector in the OON switching state, V OPO is the original negative basic small vector in the OPO switching state, V PON is the original basic vector in the PON switching state, V OPN is the original basic center vector in the OPN switching state, V PNO is the original basic center vector in the PNO switching state, V PNN is the original basic vector in the PNN switching state, V PPN It is the original basic large vector under the PPN switching state.

[0048] The step S2 adjusts the action time of the virtual basic vector according to the time correction factor through the following formula:

[0049] T′2=fT2

[0050] Wherein, f is the time correction factor, T2 is the original action time of the vector in the virtual basic, and T′2 is the action time of the vector in the virtual basic after adjustment.

[0051] The time correction factor is obtained by solving the following formula:

[0052]

[0053] Where f is the time correction factor, C is the capacitance value, ΔU is the voltage difference between the two capacitors, that is, the midpoint potential, T2 is the original action time of the virtual basic vector, i b is the load current of phase B; since the working time of the vector cannot be negative, the absolute value of the correction factor f cannot be greater than 1. When the correction factor is too large, the effect of the vector action will be weakened, resulting in distortion of the output voltage waveform. When the correction factor is too small, it will not play the role of correcting the midpoint potential.

[0054] The calculation process of the time correction factor is:

[0055] The problem of midpoint potential fluctuation is reflected in the unequal voltage of the two capacitors on the DC side. Among them, U is the voltage and Q is the charge.

[0056] The midpoint potential fluctuation means that the charge at the midpoint is not equal to 0. Therefore, keeping the charge at the midpoint at 0 can suppress the midpoint potential fluctuation and correct the deviation of the midpoint potential.

[0057] The midpoint charge ΔQ flowing out during a switching cycle * for:

[0058]

[0059] Among them, i a 、i b 、i c is the three-phase load current, and i a +i b +i c =0;

[0060] Then the midpoint charge ΔQ * Simplified to: The charge change measured in one cycle is: ΔQ = Q2-Q2 = C(U C2 -U C1 )=CΔU,where U C1 , U C2 are the voltages of the two capacitors on the DC side, and Q2 and Q1 are the charges of the two capacitors on the DC side. When the charge change in this cycle offsets the charge surplus in the previous cycle, the midpoint potential is balanced, and we have: ΔQ * +ΔQ=0,simultaneous and i a +i b +i c =0, calculated

[0061] Table 1 shows the time correction factor f corresponding to each large interval:

[0062] Table 1

[0063]

[0064] Taking the reference vector located in the 4th zone in the 1st zone as an example, the output sequence and action time of the basic vectors change as follows: Figure 5 After being restricted by the above conditions, the obtained correction factor is substituted into the vector corresponding action time to control the balance of the midpoint potential.

[0065] Figure 6As shown in (a), the midpoint potential obtained using the traditional virtual space vector algorithm will fluctuate stably in the stage, but as time increases, the midpoint potential gradually shifts and becomes unstable. Figure 6 As shown in (b), the method of the present invention ensures that the midpoint potential does not shift with time and remains stable within ±0.5V, which has a good inhibitory effect on the fluctuation of the midpoint potential. Figure 7 As shown in (b), the common mode voltage amplitude obtained by the method adopted by the present invention is U dc / 6=100V, which is lower than the common mode voltage U obtained by the traditional virtual space vector algorithm. dc / 3=200V reduced by half. Figure 8 As shown, assuming that the voltage values ​​of the two capacitors on the DC side are set to 305V and 295V respectively at the initial moment, and the midpoint potential has an initial deviation of 10V, the method adopted by the present invention realizes the control of the midpoint potential balance in about 0.03s.

Claims

1. A midpoint potential control method for adjusting the common mode voltage in a T-type three-level inverter, characterized in that: The following steps are involved: S1. Define a virtual basic vector to avoid the participation of positive vectors; S2, according to the virtual basic vector, using the space vector pulse width modulation method, and adjusting the action time of the virtual basic vector according to the time correction factor, to balance the midpoint potential of the common mode voltage in the T-type three-level inverter; The step S2 adjusts the action time of the virtual basic vector according to the time correction factor through the following formula: T′2=fT2 Among them, f is the time correction factor, T2 is the original action time of the virtual basic center vector, and T′2 is the action time of the virtual basic center vector after adjustment; The time correction factor is obtained by solving the following formula: Where f is the time correction factor, C is the capacitance value, Δu is the voltage difference between the two capacitors, that is, the midpoint potential, T2 is the original action time of the virtual basic vector, i b is the B phase load current.

2. The midpoint potential control method for adjusting the common mode voltage in a T-type three-level inverter according to claim 1, characterized in that: The definition of the virtual basic vector in step S1 includes: V N0 =V OOO V NL1 =V PNN V NL2 =V PPN Among them, V N0 is the virtual basic zero vector, V NS1 is the first virtual basic vector, V NS2 is the second virtual basic vector, V NM1 is the virtual basic vector, V NL1 is the first virtual basic vector, V NL2 is the second virtual basic vector, V OOO is the original zero vector under OOO switching states, V OON is the original negative basic small vector in the OON switching state, V POO is the original negative basic small vector in POO switching states, V ONO is the original negative basic small vector in the OON switching state, V OPO is the original negative basic small vector in the OPO switching state, V PON is the original basic vector in the PON switching state, V OPN is the original basic center vector in the OPN switching state, V PNO is the original basic center vector in the PNO switching state, V PNN is the original basic vector in the PNN switching state, V PPN It is the original basic large vector under the PPN switching state.

Images