CN109120167B

CN109120167B - Sector buffer-based two-level PWM rectifier fault-tolerant control method

Info

Publication number: CN109120167B
Application number: CN201810928516.0A
Authority: CN
Inventors: 冯延晖; 秦伟; 邱颖宁; 黄凯
Original assignee: Nanjing University of Science and Technology
Current assignee: Nanjing University of Science and Technology
Priority date: 2018-08-15
Filing date: 2018-08-15
Publication date: 2020-09-11
Anticipated expiration: 2038-08-15
Also published as: CN109120167A

Abstract

The invention discloses a sector buffer-based two-level PWM rectifier fault-tolerant control method, which selects a sector division mode to divide sectors; determining the influence of a fault switch tube on each sector and the change of basic voltage vectors before and after a fault; determining the basic voltage vector of each sector and the acting time of the basic voltage vector before the fault; selecting a normal sector, in which the reference voltage rotation vector is transited from the fault sector to the normal sector, as a buffer sector; adjusting the action time of the basic voltage vector of the sector and the buffer sector influenced by the fault switching tube; and determining the conduction time of the three-phase switch tube, modulating the conduction time of the switch tube with a triangular carrier, determining PWM (pulse-width modulation) pulse of the switch tube, determining the on-off state of the switch tube, and finishing the buffer fault-tolerant control. The invention can accurately compensate the sector influenced by the fault of the switching tube, further reduces the negative influence caused by the fault switching tube by a sector buffering method, and can improve the fault-tolerant control effect on single-tube faults and double-tube faults.

Description

Sector buffer-based two-level PWM rectifier fault-tolerant control method

Technical Field

The invention relates to a sector buffer-based two-level PWM rectifier fault-tolerant control method, and belongs to the field of power conversion and control.

Background

With the development of power electronic technology, the three-phase two-level PWM rectifier has sinusoidal input current, bidirectional energy flow and adjustable direct-current voltage, and is widely applied to the fields of medium and high-power occasions such as offshore wind power generation, new energy electric vehicles and the like. In most cases, the rectifier needs to work continuously for a long time in a severe industrial environment, and the rectifier fails inevitably due to factors such as unreliability and improper control of the power switch tube. To avoid major accidents and to reduce down time, the system must be fault-tolerant controlled to restore as much performance as possible before the failure.

The existing three-phase two-level PWM rectifier fault-tolerant modes are divided into two categories, namely hardware fault-tolerant control and software fault-tolerant control. The software fault-tolerant mode can carry out fault-tolerant processing on a fault by changing the system operation strategy and control parameters when the switching tube has the fault, does not need to change the existing hardware layout of the system and add redundant parts, and utilizes the devices which are not in fault of the original system to recover to the operation state before the fault to the maximum extent. Chinese patent 201510277790.2 proposes a fault-tolerant control method for a three-phase bridge PWM rectifier, which corrects a reference voltage vector by correcting a switching pattern, thereby realizing fault-tolerant operation of the rectifier. The method does not perform accurate compensation aiming at the influence of a fault switching tube on each sector, performs compensation on the sector without the fault of the switching tube, and belongs to overcompensation. The article "PWM rectifier fault-tolerant control system based on NCAV and circuit equivalent replacement" proposes a fault-tolerant control method of a PWM rectifier based on an equivalent circuit, which does not perform precise compensation for the influence of a faulty switching tube on each sector, and does not perform compensation in a sector where a plurality of fault vectors commonly influence, which belongs to under-compensation. A thesis 'rectifier fault-tolerant control method based on space vector control' provides a rectifier fault-tolerant control method based on space vector, a unified sector partition function is not established in the method, fault-tolerant control on faults of a plurality of bridge arm switch tubes cannot be achieved, although accurate compensation of a single switch tube can be achieved, negative effects of the fault switch tube on a system are not considered, and the effect of the fault-tolerant control is not optimal.

Disclosure of Invention

The invention aims to provide a sector buffer-based two-level PWM rectifier fault-tolerant control method, which can accurately compensate sectors affected by switching tube faults, further reduce the negative effects brought by faulty switching tubes by the sector buffer method, and improve the fault-tolerant control effect on single-tube faults and double-tube faults.

The technical solution for realizing the purpose of the invention is as follows: a sector buffer-based two-level PWM rectifier fault-tolerant control method comprises the following steps:

step 1, selecting a sector division mode to carry out sector division;

step 2, determining the influence of the fault switch tube on each sector and the change of basic voltage vectors before and after the fault according to the position of the fault switch tube;

step 3, determining the basic voltage vector of each sector and the acting time of the basic voltage vector before the fault;

step 4, selecting a normal sector of the reference voltage rotation vector which is transited from the fault sector to the normal sector as a buffer sector;

step 5, adjusting the action time of the basic voltage vectors of the sector and the buffer sector influenced by the fault switching tube according to the change of the fault basic voltage vectors and the action time of the basic voltage vectors before the fault;

step 6, determining the conduction time of the three-phase switch tube according to the action time of the basic voltage vector and the sector type;

and 7, modulating the conduction time of the switching tube with a triangular carrier, determining the PWM pulse of the switching tube, determining the on-off of the switching tube, and finishing the buffer fault-tolerant control.

Compared with the prior art, the invention has the following remarkable advantages: 1) the sector buffering method is adopted to solve the problems that the current spike and the like can occur when a current track circle transits from a fault sector to a normal sector due to the fact that the inductance of a converter circuit is increased and the capacitance in the circuit is reduced caused by the fact that the software fault-tolerant control of the two-level PWM rectifier is adopted; 2) the invention can improve the fault-tolerant control effect of single-tube fault and double-tube fault of the two-level PWM rectifier; 3) the SVPWM algorithm in the main controller is only needed to be modified, the algorithm is simple and easy to realize, and extra hardware cost is not needed.

Drawings

FIG. 1 is a control block diagram of a side converter system topology of a direct-drive wind driven generator and a fault-tolerant buffer control method thereof.

Fig. 2 is a schematic diagram of the main circuit topology of the three-phase two-level PWM rectifier of the present invention.

FIG. 3 is a flow chart of a method for improving the fault-tolerant effect of a two-level PWM rectifier based on sector buffering according to the present invention.

Fig. 4 is an eight-sector basic space voltage vector diagram in an alpha and beta two-phase stationary coordinate system under the fault-tolerant control condition of the three-phase two-level PWM rectifier of the present invention.

Fig. 5 is a twelve-sector basic space voltage vector diagram in an alpha and beta two-phase stationary coordinate system under the fault-tolerant control condition of the three-phase two-level PWM rectifier of the present invention.

Fig. 6 is a schematic diagram of the distribution of the sectors and the buffer sectors affected by the faulty switching tube when the S1 tube fault is in the eight-sector division mode.

Fig. 7 is a schematic diagram of the distribution of the sectors and the buffer sectors affected by the faulty switching tube in the twelve-sector division mode of the three-phase two-level PWM rectifier S1 according to the present invention.

Fig. 8 is a schematic diagram of the distribution of the sectors and the buffer sectors affected by the faulty switching tube in the twelve-sector division mode when the fault occurs in the two-tube fault of the three-phase two-level PWM rectifiers S1 and S3 according to the present invention.

Fig. 9 is a schematic diagram of the distribution of the sectors and the buffer sectors affected by the faulty switching tube in the twelve-sector division mode when the fault occurs in the two-tube fault of the three-phase two-level PWM rectifiers S1 and S4 according to the present invention.

Fig. 10 is a schematic diagram of the distribution of the sectors and the buffer sectors affected by the faulty switching tube in the twelve-sector division mode when the fault occurs in the two-tube fault of the three-phase two-level PWM rectifiers S1 and S6 according to the present invention.

Fig. 11 is a schematic distribution diagram of the sectors and buffer sectors affected by the faulty switching tube in the twelve-sector division mode of the three-phase two-level PWM rectifiers S1, S3 and S5 according to the present invention.

Fig. 12 is a sector viii voltage vector composite diagram of the three-phase two-level PWM rectifier S1 tube fault in twelve sector division mode according to the present invention.

FIG. 13 is a vector diagram of the buffered sector XI voltage vector combination for the three-phase two-level PWM rectifier S1 in twelve-sector partitioning mode.

FIG. 14 is a diagram of a three-phase two-level PWM rectifier S1 according to the present invention, which generates PWM in one cycle with normal and buffered sectors.

Fig. 15 is a current vector locus diagram of an alpha and beta two-phase stationary coordinate system in three states of fault-tolerant control, single-sector fault-tolerant buffer control and double-sector fault-tolerant buffer control in a twelve-sector division mode for a three-phase two-level PWM rectifier S1 according to the present invention.

Fig. 16 is a current vector trajectory diagram of an alpha and beta two-phase stationary coordinate system in three states of fault-tolerant control, single-sector fault-tolerant buffer control and double-sector fault-tolerant buffer control in a twelve-sector division mode for double-tube faults of the three-phase two-level PWM rectifiers S1 and S3 according to the invention.

FIG. 17 is a waveform diagram of the rotation speed of the generator under five states of normal operation, fault-tolerant operation, single-sector fault-tolerant buffer control and double-sector fault-tolerant buffer control of the three-phase two-level PWM rectifier S1 according to the present invention.

The reference numbers in the figures illustrate: 6 power switching tubes in S1-S6 three-phase two-level PWM rectifier, 6 fly-wheel diodes in D1-D6 three-phase two-level PWM rectifier, 6 thermal fuses in F1-F6 three-phase two-level PWM rectifier, and L_a,L_b,L_cFor generator stator winding equivalent inductance, U_a,U_b,U_cIs the equivalent voltage source of the generator. C is a direct current side voltage stabilizing capacitor. Three-phase current i_a,i_b,i_cThree-phase current, v, generated for a permanent-magnet synchronous generator_wIs the magnitude of natural wind speed, omega_mIs the angular velocity of the permanent magnet synchronous generator, theta is the three-phase current electrical angle, i_d,i_qA given value of current under a dq two-phase rotating coordinate system,

is a current feedback value under a dq two-phase rotating coordinate system,

is a feedback value of the torque of the motor,

t is a reference voltage component under α, β two-phase stationary coordinate system_fAt the moment of failure of the switching tube, t_comThe moment of fault-tolerant control of the fault. t is t_com1The time for single-sector fault-tolerant buffer control of the fault. t is t_com2The time when the fault is subjected to double-sector fault-tolerant buffer control.

Detailed Description

The invention is further illustrated by the following examples in conjunction with the accompanying drawings.

FIG. 1 is a control block diagram of a machine side converter system topology of a permanent magnet direct-drive wind driven generator and a fault tolerance control method thereof, and FIG. 2 shows a three-phase two-level control methodThe main circuit topology schematic diagram of the PWM rectifier is equivalent to a voltage source U on the generator side_a,U_b,U_cAnd stator inductance L_a,L_b,L_c. In practical application, the probability that the power switch tube and the diode connected in anti-parallel with the power switch tube simultaneously have faults is extremely low, so that the power switch tube and the diode connected in anti-parallel with the power switch tube are still normally operated by default only considering that the power switch tube has faults. When the power switch tube in fig. 2 (S)₁～S₆) Faults, mainly open faults and short faults, the short faults being caused by thermal fuses (F) connected in series₁～F₆) And converting the fault into an open-circuit fault, wherein other faults can be combined with a fault diagnosis algorithm to convert the fault into the open-circuit fault by closing a fault switching tube to drive a pulse signal. In summary, the invention provides a sector buffer based two-level PWM rectifier fault-tolerant control method, as shown in fig. 3, the steps are as follows:

step 1, selecting a sector division mode to carry out sector division, wherein the sector division mode comprises an eight sector division mode and a twelve sector division mode, and the specific process of the two sector division modes is as follows:

(1) if the eight-sector division method is adopted, a symbol function is defined:

wherein i ═ a, B, C, D, E, F;

as the faults of the three bridge arms of a, b and c correspond to different sector division coordinate systems, in order to accurately carry out fault-tolerant control on fault sectors, a function N of the three sector division coordinate systems is defined_a,N_b,N_c。

Let N_a＝H(A)+H(B)+4H(C)+3H(D)

Let N_b＝4G(B)+3G(C)+G(D)+H(E)

Let N_c＝3H(B)+H(C)+4H(D)+H(F)

Determination of the calculated value N from Table 1_a,N_b,N_cCorresponding relation with actual sector number

TABLE 1 calculation of value N_a,N_b,N_cCorresponding relation with sector

Sector numbering	I	Ⅱ	Ⅲ	Ⅳ	Ⅴ	Ⅳ	Ⅶ	Ⅷ
									Calculating the value N _a	6	2	4	3	7	5	1	8
Calculating the value N _b	2	4	3	7	5	6	1	8
									Calculating the value N _c	4	3	7	5	6	2	1	8

According to reference voltage components of α and β in two-phase stationary coordinate system

Determining a rotating reference vector V_ref ^*When rotating the reference vector V_ref ^*Rotating for one circle to obtain a calculated value N_aThe sequence of change of (A) is: 6 → 2 → 1 → 4 → 3 → 7 → 8; calculating the value N_bThe sequence of change of (2 → 1 → 4 → 3 → 7 → 8 → 5 → 6; calculating the value N_cThe sequence of change of (a) is 4 → 3 → 7 → 8 → 5 → 6 → 2 → 1;

selecting different sector division functions according to the positions of the fault switching tubes, and selecting the sector division function N when the a-phase bridge arm switching tube has a fault_aThrough N_aThe change sequence of the actual sector numbers is determined, that is, the eight-sector division is as shown in fig. 4 (a); when a b-phase bridge arm switching tube fault selects a sector division function N_bThrough N_bThe change sequence of the actual sector numbers is determined, that is, the division of eight sectors is shown in fig. 4 (b); when a c-phase bridge arm switching tube fault selects a sector division function N_cThrough N_cDetermines the actual sectorThe changing sequence of the numbers, namely the division of eight sectors is shown in fig. 4 (c);

(2) if a twelve-sector division method is adopted, a symbol function is defined:

wherein i is a, B, C, D, E, F.

Let N ═ sign (A) + sign (B) +2sign (C) +2sign (D) +4sign (E) +3sign (F)

Determining the corresponding relation between the calculated value N and the actual sector number through the table 2;

TABLE 2 calculated value N and sector corresponding relation

Calculating the value N	1	2	3	4	5	6	7	8	9	10	11	12
													Sector numbering	Ⅷ	Ⅱ	Ⅲ	Ⅶ	Ⅳ	Ⅸ	Ⅻ	I	Ⅹ	Ⅵ	Ⅴ	Ⅺ

Determining a rotating reference vector V_ref ^*When rotating the reference vector V_ref ^*After one rotation, the change sequence of the calculated values N is as follows: 8 → 4 → 2 → 1 → 3 → 6 → 5 → 9 → 11 → 12 → 10 → 7 → 8, i.e. the order of change of the actual sector numbers, i.e. the division of the twelve sectors is shown in FIG. 5.

the method for determining the influence of a fault switching tube on each sector and dividing the fault sector and the normal sector comprises the following steps:

if an eight-sector division mode is adopted, determining the sectors affected by the fault switching tube in the eight sectors, namely fault sectors, according to the tables 3-5; if a twelve-sector division mode is adopted, determining the sector influenced by the fault switch tube in the twelve sectors according to the table 6;

TABLE 3 affected sectors corresponding to eight-sector a-phase bridge arm switching tube fault

Table 4 affected sectors corresponding to eight-sector b-phase bridge arm switching tube fault

TABLE 5 affected sectors corresponding to eight-sector c-phase bridge arm switching tube faults

TABLE 6 affected sectors corresponding to twelve-sector single switch tube failure

In the table, the gray part indicates that the sector is affected by the faulty switching tube, i.e. the faulty sector, and the white part indicates that the sector is not affected by the faulty switching tube, i.e. the normal sector.

The method for determining the basic voltage vectors before and after the fault of the switching tube comprises the following steps: determining the change conditions of basic voltage vectors, namely a fault zero vector and an effective vector before and after the fault of the switching tube according to the table 7, and determining a fault voltage vector;

TABLE 7 Voltage vector variation before and after single switch tube failure

In the table, the upper and lower switching states of the same bridge arm are set to be complementary, that is, the upper bridge arm of the same bridge arm is turned on, the lower bridge arm must be turned off, the state is recorded as 1, similarly, the upper bridge arm of the same bridge arm is turned off, the lower bridge arm is turned on, the state is recorded as 0, and under the premise that the three bridge arms are complementary to the same bridge arm, 8 switching state combinations are provided, wherein "000", "100", "110", "010", "011", "001", "101" and "111" correspond to eight basic voltage vectors, and the eight basic voltage vectors include six effective vectors and two zero vectors.

And 3, selecting a normal sector with the reference voltage rotation vector transited from the fault sector to the normal sector as a buffer sector, wherein the specific method comprises the following steps: the number of buffer sectors is selected according to the actual requirement, and 1 or two continuous sectors adjacent to the fault sector are selected as the buffer sectors, as shown in FIGS. 6-8.

Step 4, determining the basic voltage vector of each sector and the acting time of the basic voltage vector before the fault, wherein the specific method comprises the following steps:

first, the intermediate variables are defined as:

in the formula (I), the compound is shown in the specification,

is a reference voltage component in α, β two-phase stationary coordinate system, U_dcFor the output voltage of the DC side, T_sIs a sampling period;

then, determining the action time T of the effective vector in the basic voltage vector of each sector₁And T₂；

If the voltage vector is eight sectors, determining the action time T of the effective vector in the basic voltage vector of each sector according to tables 8-10₁And T₂；

TABLE 8 a relationship between the action time of the sector and the basic voltage vector during the phase bridge arm failure

TABLE 9 b relationship between the action time of the sector and the basic voltage vector during the phase bridge arm failure

TABLE 10 c relationship between the action time of the sector and the basic voltage vector during the phase bridge arm failure

If the voltage vector is twelve sectors, determining the action time T of the effective vector in the basic voltage vector of each sector according to the table 11₁And T₂；

TABLE 11 sector vs. base Voltage vector action time relationship

Then, the effective vector is applied for a time T₁And T₂Calculating the action time T of the zero vector in the basic voltage vector₀＝T_s-T₁-T₂；

Finally, overmodulation judgment is carried out, namely whether the sum of two times is greater than a sampling period is judged after the action time of two non-zero basic voltage vectors is calculated, if the sum of the two times is greater than the sampling period, the output voltage is seriously distorted, the two times need to be redistributed, and the distribution principle is as follows:

the resultant reference voltage rotation vector V_ref ^*The proportionality coefficients are:

namely:

the specific method for adjusting the action time of the basic voltage vector of the sector influenced by the fault switch tube comprises the following steps:

(1) adopts an eight-sector division mode

For the sector only affected by the zero vector and without simultaneous fault of the zero vector, the normal zero vector is used to replace the fault zero vector, i.e. the action time of the normal zero vector is set as T₀Realizing the fault-tolerant control of the sector;

for the sector which is affected by a plurality of fault voltage vectors and has no simultaneous fault of zero vectors, the normal zero vector is used for replacing the fault zero vector to complete the compensation of the zero vector, the effective vector without fault is used, the effective vector action time is calculated based on the compensation principle to synthesize the reference voltage rotation vector V again_ref ^*Realizing the fault-tolerant control of the sector, wherein the compensation principle comprises a mapping method, an equiaxial component method and an equiaxial method;

(2) adopts a twelve-sector division mode

For the sector only affected by the zero vector and without simultaneous fault of the zero vector, the normal zero vector is used to replace the fault zero vector, i.e. the action time of the normal zero vector is set as T₀Implementing fault-tolerant control of the sector, T₀Acting time of zero vector before fault;

for the sectors which are affected by a plurality of fault voltage vectors and have no simultaneous fault in the zero vector, the normal zero vector is used for replacing the fault zero vector, the effective vector without the fault is used, the effective vector action time is calculated based on the compensation principle to synthesize the reference voltage rotation vector V again_ref ^*To realizeFault-tolerant control of the sector, the compensation principle includes mapping method, equiaxial component method and equiaxial method;

for the sector with simultaneous fault of zero vectors, because no normal zero vector exists in the sector, the output vector can not be adjusted, and the reference voltage rotates the vector V_ref ^*The output module value reaches the maximum, and the sector can not carry out fault-tolerant control;

for the sector which is affected by a plurality of fault vectors and has a fault at the same time, because the sector has no normal zero vector, the output vector can not be adjusted, and the reference voltage rotates the vector V_ref ^*The output modulus reaches the maximum, and the sector can not carry out fault-tolerant control.

Each compensation time determination method is described in detail below.

The mapping method is a method commonly used in the prior art, and is not described herein again.

The normal mode method is to make vector V before fault_ref ^*The compensation time of the effective vector is calculated in conformity with the modulo length of the effective vector. Namely, the normal zero vector is used to replace the fault zero vector to complete the compensation of the zero vector, and the effective vector without fault is used to rotate the reference voltage by a vector V_ref ^*Orthogonally mapped to the effective vector, rotating the vector V based on the reference voltage_ref ^*Calculating action time of normal effective vector by using equal modulus principle

Implementing fault tolerant control of the sector.

The equiaxed component method is the vector V to be before failure_ref ^*And the effective vector is projected on an β axis, the effective vector compensation time is calculated based on the component identity, namely, a normal zero vector is used for replacing a fault zero vector, and the reference voltage is rotated by a vector V_ref ^*Projected on an β axis, calculating the compensation ratio of the normal effective voltage vector based on the principle of equal β axis components, namely setting the acting time T of the normal effective vector₁+T₂Implementing fault-tolerant control of the sector, T₁And T₂For making two effective vectors before faultThe time is used.

The specific method for adjusting the action time of the basic voltage vector of the buffer sector comprises the following steps: the defects of increase of inductance of a converter circuit, reduction of capacitance in the circuit and the like caused by reduction are overcome, and the problems of current spike, large rotation speed fluctuation and the like caused when a current track circle is transited from a fault sector to a normal sector are prevented. And controlling the vector action time when the fault zero vector is normalized in the buffer sector, replacing the zero vector of the periodic fault with the zero vector which does not have the fault in the buffer sector, ensuring that the reference voltage rotating vectors in the buffer sector before and after the change are consistent, and calculating the compensation time of the effective vector.

And 6, determining the conduction time of the three-phase switch tube according to the action time and the sector type of the basic voltage vector, wherein the specific method comprises the following steps:

(1) adopts an eight-sector division mode

Preferably, for a sector which is not affected by a fault vector, the conduction time of a three-phase switch tube is defined as:

in the formula, T_sFor a sampling period, T₁And T₂The action time of the effective vector;

at this time, the reference voltage rotates the vector

Taking the vector synthesis of the voltage of the sector VII under the twelve-sector division function of the S1 pipe fault as an example, as shown in FIG. 12, the implementation of the continuous pulse width modulation on the reference voltage V is realized_ref ^*And (4) synthesizing.

Defective sector by redefining defective sector T_a,T_b,T_cTime variable, compensation of fault vector in each fault sector, i.e. replacement of fault vector based on specific equivalence principle by using effective vector, rotating reference voltage by vector V_ref ^*The method can be used for re-synthesizing or approximately recovering in a fault sector, and comprises the following specific steps:

for the sector which is only affected by the zero vector and has no simultaneous fault of the zero vector, the conduction time of the three-phase switch tube is changed only in the sector affected by the zero vector, and the T is redefined_a,T_b,T_c；

When the position of the fault switch tube is the upper bridge arm, redefining as:

when the position of the fault switch tube is the lower bridge arm, redefining as:

for the sector affected by the effective vector and without simultaneous fault of the zero vector, the conducting time of the three-phase switch tube is changed only in the sector affected by the zero vector, and the action time calculated according to the compensation principle redefines T_a,T_b,T_cThe compensation principle is different from the definition formula.

(a) The mapping method is a method commonly used in the prior art, and how to redefine T is not described herein again_a,T_b,T_c。

(b) For the isocode method: when the position of the fault switch tube is the upper bridge arm, redefining as:

(c) for the equiaxed component method: when the position of the fault switch tube is the upper bridge arm, redefining as:

the buffer sector can realize the replacement of the normalized zero vector by the normal zero vector by redefining the time variable, and the reference voltage rotating vector is

The reference voltage rotation vector after the fault-tolerant buffer control is

The consistency of the reference voltage rotation vector can be ensured. Taking the vector synthesis of the voltage of the sector XI under the twelve-sector partition function of the S1 tube fault as an example, as shown in FIG. 13, the fault-tolerant buffer control of the S1 tube fault is realized by the specific method: when the position of the fault switch tube is the upper bridge arm, redefining as:

then, determining the switching time of each sector according to tables 12-14;

TABLE 12 relationship between the action time of the sector and the basic voltage vector when the eight-sector a-phase bridge arm fails

TABLE 13 relationship between the action time of the sector and the basic voltage vector when the eight-sector b-phase bridge arm fails

TABLE 14 relationship between the action time of the sectors and the basic voltage vector during eight-sector c-phase bridge arm failure

(2) Adopts a twelve-sector division mode

for the sectors which are commonly influenced by a plurality of fault voltage vectors and have no simultaneous fault of the zero vector, three sectors need to be changed only in the sectors influenced by the zero vectorRedefining T according to action time calculated by compensation principle_a,T_b,T_cDifferent definition formulas of the compensation principle are different, and the compensation principle is specifically the same as the calculation formula of the eight-sector type sector;

for the buffer sector, the calculation formula is the same as that of the eight-sector type sector;

then, determining the switching time of each sector according to the table 15;

TABLE 15 on-off time distribution relationship of different twelve sectors

Sector numbering

I

Ⅱ

Ⅲ

Ⅳ

Ⅴ

Ⅵ

Ⅶ

Ⅷ

Ⅸ

Ⅹ

Ⅺ

Ⅻ

Conduction time T of A-phase switch tube_a

T_a

T_b

T_c

T_b

T_a

T_b

T_c

T_b

T_a

Conduction time T of B-phase switch tube_b

T_b

T_a

T_b

T_c

T_b

T_a

T_b

T_c

Conduction time T of C-phase switch tube_c

T_c

T_b

T_a

T_b

T_c

T_b

T_a

T_b

And 7, modulating the conduction time of the switching tube with a triangular carrier, determining PWM (pulse-width modulation) pulse of the switching tube, determining the on-off state of the switching tube, and finishing buffer fault-tolerant control, wherein the specific method comprises the following steps of: the method comprises the steps of modulating isosceles triangle waves with the switch tube conduction time and period as sampling periods, determining the action sequence of vectors by adopting a DWPM technology based on a symmetry principle and a THD minimum principle to obtain 6 paths of PWM pulses, acting the output 6 paths of PWM pulses on a power switch tube driving circuit, and driving the corresponding power switch tubes to be switched on and off by the driving circuit to complete fault-tolerant buffer control.

By adopting the technical scheme, the invention can solve the problems of current spike, generator rotating speed fluctuation and the like when a current track circle is transited from a fault sector to a normal sector due to the fact that the inductance of a converter circuit is increased and the capacitance in the circuit is reduced caused by software fault-tolerant control of the three-phase two-level PWM rectifier by providing the sector buffer method, can realize the single-tube fault and double-tube fault-tolerant control effect of the two-level PWM rectifier, and reduces the influence of a fault switching tube on a system. The scheme is simple and easy to realize through a reconstruction controller SVPWM algorithm, and extra hardware cost is not required to be increased.

To verify the effectiveness of the present invention, the following simulations were performed.

Example 1

In this embodiment, a single-tube fault S1 of a three-phase two-level PWM rectifier is illustrated in an eight-sector division manner, when a short-circuit fault occurs in the S1 tube, the single-tube fault is converted into an open-circuit fault by a thermal fuse, and when an open-circuit fault occurs in the S1 tube, a fault-tolerant buffer control method is illustrated according to table 16 comparing sector variation conditions in three states of fault-tolerant control, single-sector fault-tolerant buffer control and double-sector fault-tolerant buffer control single-cycle.

TABLE 16S 1 comparison of eight-sector single-cycle three-state sectors under switching tube failure

The distribution schematic diagram of the sectors and buffer sectors affected by the fault switch tube when the S1 tube fault of the three-phase two-level PWM rectifier is in the eight-sector division mode is shown in FIG. 6, under the condition of the S1 tube fault, the normal ground reference voltage rotating vector is controlled for the sectors not affected by the fault switch tube in the table 16, and the reference voltage V is modulated by continuous pulse width through a seven-segment switching sequence_ref ^*And (4) synthesizing. For the sectors affected by the faulty switch tube in the table 16, software fault-tolerant control is performed by redefining the faulty sector T_a，T_b，T_cTime variable, compensation of fault vector in each fault sector, i.e. replacement of fault vector based on specific equivalence principle by using effective vector, rotating reference voltage by vector V_ref ^*It can be resynthesized or approximately recovered in the failed sector. For the buffer sectors in Table 16, T is redefined within the buffer sector_a，T_b， T_cA time variable. Controlling fault vector V in buffer sector₀Normalized action time using non-faulted zero vector V₇Zero vector V instead of periodic faults₀Therefore, fault-tolerant buffer control of single tube fault of the three-phase two-level PWM rectifier S1 under the eight-sector division mode is realized, and when other single switching tubes are in fault, the fault-tolerant buffer control method is adopted for capacity toleranceAnd (4) error.

Example 2

The present embodiment is illustrated with a three-phase two-level PWM rectifier with multiple tube faults in an eight-sector division.

When two switching tubes in a three-phase two-level PWM rectifier fail simultaneously, as shown in fig. 2, the following four situations are divided:

a. two upper and lower switching tubes of the same bridge arm simultaneously break down;

b. two upper pipes of different bridge arms simultaneously break down;

c. two lower tubes of different bridge arms simultaneously break down;

d. one upper pipe and one lower pipe of different bridge arms are in failure.

Because the short-circuit fault can be converted into the open-circuit fault through the thermal fuse, the fault of the switching tube can be converted into the open-circuit fault through the turn-off controller after other fault diagnoses, and the open-circuit fault is explained here.

In case a, the case where the a-phase arm upper and lower switching tubes S1 and S4 fail at the same time will be described. Under the eight-sector division mode, the table 1 shows that the sectors affected by the faults of the switching tubes S1 and S4 are complementary, the action polarity ranges of the phase-A current are also complementary, and the situation can be regarded as two single-tube faults to carry out fault-tolerant control. But the fault switch tube affects all sectors, and the normal sector can not be used as a buffer sector to realize fault-tolerant buffer control.

For the conditions b, c and d, because three different eight-sector division functions are adopted for the faults of the switching tubes of the bridge arms of the phase a, the phase b and the phase c, the faults all relate to the fault-tolerant control of different bridge arms, and the fault-tolerant control and the fault-tolerant buffer control cannot be carried out.

When three or more switching tubes simultaneously break down, the failures all relate to fault-tolerant control of different bridge arms due to three different eight-sector division functions adopted by the failures of the switching tubes of the a-phase bridge arm, the b-phase bridge arm and the c-phase bridge arm, and the failures cannot be subjected to fault-tolerant control or fault-tolerant buffer control.

Example 3

In this embodiment, a single-tube fault S1 of a three-phase two-level PWM rectifier is illustrated in a twelve-sector division manner, when a short-circuit fault occurs in the S1 tube, the single-tube fault is converted into an open-circuit fault by a thermal fuse, and when an open-circuit fault occurs in the S1 tube, a fault-tolerant buffer control method is illustrated according to table 17 comparing sector variation conditions in three states of fault-tolerant control, single-sector fault-tolerant buffer control and double-sector fault-tolerant buffer control single-cycle.

TABLE 17S 1 twelve-sector single-cycle three-state sector comparison under switching tube failure

The distribution schematic diagram of the sectors and buffer sectors affected by the fault switch tube when the S1 tube fault of the three-phase two-level PWM rectifier is in the twelve-sector division mode is shown in FIG. 7, under the condition of the S1 tube fault, the normal ground reference voltage rotating vector is controlled for the sectors not affected by the fault switch tube in the table 16, and the reference voltage V is modulated by continuous pulse width through a seven-segment switching sequence_ref ^*And (4) synthesizing. For the sector affected by the fault switch tube in the table 17, the software fault-tolerant control is carried out, and the fault sector T is redefined_a，T_b，T_cTime variable, compensation of fault vector in each fault sector, i.e. replacement of fault vector based on specific equivalence principle by using effective vector, rotating reference voltage by vector V_ref ^*It can be resynthesized or approximately recovered in the failed sector. For the buffer sectors in Table 17, T is redefined within the buffer sector_a，T_b，T_cA time variable. Controlling fault vector V in buffer sector₀Normalized action time using non-faulted zero vector V₇Zero vector V instead of periodic faults₀Thereby realizing the fault-tolerant buffer control of the single tube fault of the three-phase two-level PWM rectifier S1 in the twelve-sector division mode, and fig. 14 shows that the three-phase two-level PWM rectifierThe PWM rectifier S1 tube fault-tolerant control, single sector fault-tolerant buffer control and double sector fault-tolerant buffer control are in three states, namely a current vector locus diagram of α and β two-phase static coordinate system in a twelve-sector division mode, and FIG. 17 is a motor rotating speed waveform diagram in five states, namely normal operation, fault-tolerant operation, single sector fault-tolerant buffer control and double sector fault-tolerant buffer control, of the three-phase two-level PWM rectifier S1 tube of the invention.

Example 3

The present embodiment is illustrated with a two-transistor fault with a three-phase two-level PWM rectifier in a twelve-sector division.

a. two upper pipes of different bridge arms simultaneously break down;

b. two lower tubes of different bridge arms simultaneously break down;

c. two upper and lower switching tubes of the same bridge arm simultaneously break down;

d. one upper pipe and one lower pipe of different bridge arms are in failure.

In case a, a case where the tubes S1 and S3 in the two arms of a phase and B phase fail at the same time will be described. Fig. 8 is a schematic diagram showing distribution of sectors and buffer sectors affected by a faulty switching tube when a fault occurs in a two-transistor fault of the three-phase two-level PWM rectifier S1 and S3 in a twelve-sector division mode, it can be seen from fig. 8 that the sectors affected by the fault occurs in the switching tube of S1 and S3 are partially overlapped, each sector is analyzed according to the sector, and a fault-tolerant buffer control method is explained according to table 18 comparing the variation of the sectors in three states of fault-tolerant control, single-sector fault-tolerant buffer control and single-cycle double-sector fault-tolerant buffer control.

TABLE 18 twelve-sector single-cycle three-state sector comparison under double-tube fault of S1 and S3

The distribution schematic diagram of the sector and the buffer sector influenced by the fault switch tube in the twelve-sector division mode of the double-tube faults of the three-phase two-level PWM rectifiers S1 and S3 is shown in FIG. 8, under the condition of the double-tube faults of S1 and S3, the normal ground reference voltage rotating vector of the sector which is not influenced by the fault switch tube in the table 18 is controlled, and the reference voltage V is modulated by continuous pulse width through a seven-segment switching sequence_ref ^*And (4) synthesizing. For the sector affected by the fault switch tube in the table 18, the software fault-tolerant control is carried out, and the fault sector T is redefined_a，T_b，T_cTime variable, compensation of fault vector in each fault sector, i.e. replacement of fault vector based on specific equivalence principle by using effective vector, rotating reference voltage by vector V_ref ^*It can be resynthesized or approximately recovered in the failed sector. For the buffer sectors in Table 18, T is redefined within the buffer sector_a，T_b，T_cA time variable. Controlling fault vector V in buffer sector₀Normalized action time using non-faulted zero vector V₇Zero vector V instead of periodic faults₀Thus, fault-tolerant buffer control of double-tube faults of S1 and S3 of the three-phase two-level PWM rectifier under a twelve-sector division mode is realized, fig. 16 is a current vector locus diagram of α and β two-phase stationary coordinate systems under three states of fault-tolerant control, single-sector fault-tolerant buffer control and double-sector fault-tolerant buffer control of the two-tube faults of the three-phase two-level PWM rectifier S1 and S3 of the invention under a twelve-sector division mode, and it can be seen that the adoption of the capacity-tolerant two-level PWM rectifierThe error buffer control algorithm can further reduce the influence of a fault switching tube on a system after the fault-tolerant algorithm is implemented. When two upper pipes of different bridge arms simultaneously fail, the fault-tolerant buffer control method is adopted for fault tolerance.

In case B, a case where the two arm lower tubes S4 and S6 of the a-phase and the B-phase fail at the same time will be described. It can be known from table 4 that the sectors affected by the switching tube faults of S4 and S6 are partially overlapped, each sector is analyzed according to the sector, and the fault-tolerant buffer control method is explained according to the sector change conditions under three states of comparing the fault-tolerant control, the single-sector fault-tolerant buffer control and the double-sector fault-tolerant buffer control in a single period according to table 19.

TABLE 19 twelve-sector single-cycle three-state sector comparison under S4, S6 double-tube fault

The sector influenced by the fault switch tube in the twelve-sector division mode of the double-tube fault of the three-phase two-level PWM rectifier S4 and S6 is shown in table 4, under the condition of the double-tube fault of the S4 and S6, the normal ground reference voltage rotating vector is controlled for the sector which is not influenced by the fault switch tube in the table 19, and the reference voltage V is modulated by continuous pulse width through a seven-segment switching sequence_ref ^*And (4) synthesizing. For the sectors affected by the failed switch in the table 19, software fault-tolerant control is performed by redefining the failed sector T_a，T_b，T_cTime variable, compensation of fault vector in each fault sector, i.e. replacement of fault vector based on specific equivalence principle by using effective vector, rotating reference voltage by vector V_ref ^*It can be resynthesized or approximately recovered in the failed sector. For the buffer sectors in Table 19, T is redefined within the buffer sector_a，T_b，T_cA time variable. Controlling fault vector V in buffer sector₇Normalized action time using non-faulted zero vector V₀Zero vector V instead of periodic faults₇Therefore, fault-tolerant buffer control of double-tube faults of the three-phase two-level PWM rectifier S4 and S6 under a twelve-sector division mode is achieved, and fault tolerance is achieved by the fault-tolerant buffer control method when two lower tubes of different bridge arms simultaneously break down.

In case c, the case where the a-phase arm upper and lower switching tubes S1 and S4 fail at the same time will be described. It can be known from table 4 that the sectors affected by the S1 and S4 switching tube faults are complementary, and the acting polarity range of the a-phase current is also complementary, and this condition can be regarded as two single-tube faults to perform fault-tolerant control, and the fault-tolerant buffer control method is explained according to the sector change conditions in three states of table 20, namely comparing fault-tolerant control, single-sector fault-tolerant buffer control and double-sector fault-tolerant buffer control in a single period.

TABLE 20 twelve-sector single-cycle three-state sector comparison under S1, S4 double-tube fault

FIG. 9 is a schematic diagram showing the distribution of the sectors and buffer sectors affected by the faulty switch tube in the twelve-sector division mode of the two-tube fault of the three-phase two-level PWM rectifiers S1 and S4 according to the present invention, wherein in the case of the two-tube fault of S1 and S3, it can be known from the table 20 and FIG. 9 that all the sectors are faulty, the fault-tolerant control is performed on the sector affected by the faulty switch tube in the table 20, and the fault sector T is redefined_a，T_b，T_cTime variable, compensation of fault vector in each fault sector, i.e. replacement of fault vector based on specific equivalence principle by using effective vector, rotating reference voltage by vector V_ref ^*It can be resynthesized or approximately recovered in the failed sector. The fault tolerance method is adopted to carry out fault tolerance when two upper and lower switching tubes of the same bridge arm simultaneously break down, and the fault tolerance method is not adoptedIf there is a normal sector, the fault-tolerant buffer control cannot be performed.

In case d, a description will be given of a case where the a-phase upper arm switching tube S1 and the B-phase lower arm switching tube S6 both fail. It can be known from table 4 that the sectors affected by the switching tube faults of S1 and S6 are partially overlapped, each sector is analyzed according to the sector, and the fault-tolerant buffer control method is explained according to the sector change conditions in three states of comparing fault-tolerant control, single-sector fault-tolerant buffer control and double-sector fault-tolerant buffer control in a single period according to table 21.

TABLE 21 twelve-sector single-cycle three-state sector comparison under double-tube fault of S1 and S6

The distribution schematic diagram of the sector and the buffer sector influenced by the fault switch tube in the twelve-sector division mode of the double-tube faults of the three-phase two-level PWM rectifiers S1 and S6 is shown in FIG. 10, under the condition of the double-tube faults of S1 and S6, the normal ground reference voltage rotating vector of the sector which is not influenced by the fault switch tube in the table 21 is controlled, and the reference voltage V is modulated by continuous pulse width through a seven-segment switching sequence_ref ^*And (4) synthesizing. For the sectors affected by the faulty switch tube and having at least one undistorted zero vector in the table 21, software fault-tolerant control is performed by redefining the faulty sector T_a，T_b，T_cTime variable, compensation of fault vector in each fault sector, i.e. replacement of fault vector based on specific equivalence principle by using effective vector, rotating reference voltage by vector V_ref ^*It can be resynthesized or approximately recovered in the failed sector. For sectors (such as sector III, sector IV and the like) affected by two switching tube faults at the same time, zero vector V is simultaneously lacked₀、 V₇The duty ratio cannot be adjusted, and the reference voltage rotation vector V cannot be recovered_ref ^*Fault-tolerant control cannot be performed in these sectors. For the buffer sectors in Table 21, T is redefined within the buffer sector_a，T_b，T_cA time variable. Controlling fault vector V in buffer sector₀Normalized action time using non-faulted zero vector V₇Zero vector V instead of periodic faults₀Therefore, fault-tolerant buffer control of double-tube faults of the three-phase two-level PWM rectifier S1 and S6 under a twelve-sector division mode is achieved, and fault tolerance is achieved by the fault-tolerant buffer control method when one upper tube and one lower tube of different bridge arms are in fault. Because fault-tolerant control cannot be carried out on part of the sectors, the fault-tolerant control effect is poor under the condition, and the influence of the fault-tolerant buffer control on a fault switch tube is not obviously improved.

Example 4

The present embodiment is illustrated with a three-phase two-level PWM rectifier with three-transistor failure in a twelve-sector division.

When three switching tubes in a three-phase two-level PWM rectifier fail simultaneously, as shown in fig. 2, the following four situations are divided:

a. the upper tubes of the three bridge arms of the phase A, the phase B and the phase C simultaneously break down;

b. the lower tubes of the three bridge arms of the phase A, the phase B and the phase C simultaneously break down;

c. two upper tubes and one lower tube of different bridge arms have faults.

d. One upper pipe and two lower pipes of different bridge arms are in failure.

In case a, the case where the three bridge arms S1, S3, and S5 of phase a, phase B, and phase C fail at the same time will be described. Fig. 11 is a schematic diagram showing the distribution of sectors and buffer sectors affected by a fault switching tube in a twelve-sector division mode by three-tube faults of the three-phase two-level PWM rectifiers S1, S3 and S5, it can be known from fig. 11 that the sectors affected by the faults of the switching tubes of S1, S3 and S5 are partially overlapped, each sector is analyzed according to the sector, and the fault-tolerant buffer control method is explained by comparing the variation conditions of the sectors in three states of fault-tolerant control, single-sector fault-tolerant buffer control and double-sector fault-tolerant buffer control single period according to the table 22.

Twelve-sector single-cycle three-state sector comparison under three-pipe fault of tables 22S 1, S3 and S5

FIG. 11 is a schematic diagram showing the distribution of the sectors and buffer sectors affected by the faulty switching tube in the twelve-sector division mode of the three-phase two-level PWM rectifiers S1, S3 and S5 of the present invention, wherein in the case of the three-tube fault of S1, S3 and S5, it can be known from Table 22 and FIG. 11 that all the sectors are faulty, and for the sector affected by the faulty switching tube in Table 23, software fault-tolerant control is performed, and by redefining the faulty sector T_a，T_b，T_cTime variable, compensation of fault vector in each fault sector, i.e. replacement of fault vector based on specific equivalence principle by using effective vector, rotating reference voltage by vector V_ref ^*It can be resynthesized or approximately recovered in the failed sector. The fault-tolerant method is adopted for fault tolerance when the three bridge arms of the phase A, the phase B and the phase C have faults at the same time, and the fault-tolerant buffer control cannot be carried out because no normal sector exists.

For the case B, the simultaneous failure of the lower tubes S4, S6 and S2 of the three bridge arms of the phase a, the phase B and the phase C is used for explaining, it can be known from table 4 that the sectors affected by the simultaneous failure of the switching tubes of the phase S4, the phase S6 and the phase S2 are partially overlapped, each sector is analyzed according to the sector, and the fault-tolerant method is explained by comparing the normal condition, the failure condition and the change condition of the zero vector and the effective vector after the fault-tolerant control of the switching tubes according to table 23.

Twelve-sector single-cycle three-state sector comparison under three-pipe fault of tables 23S 4, S6 and S2

Under the condition of three-tube faults of S4, S6 and S2, the table 23 shows that all sectors have faults, the sectors affected by the fault switching tube in the table 23 are subjected to software fault-tolerant control, and the fault sectors T are redefined_a，T_b，T_cTime variable, compensation of fault vector in each fault sector, i.e. replacement of fault vector based on specific equivalence principle by using effective vector, rotating reference voltage by vector V_ref ^*It can be resynthesized or approximately recovered in the failed sector. The fault-tolerant method is adopted for fault tolerance when the lower tubes of the three bridge arms of the phase A, the phase B and the phase C fail simultaneously, and the fault-tolerant buffer control cannot be carried out due to the absence of normal sectors.

For the cases c and d, it can be known from table 4 that the sectors affected by the simultaneous failure of the three switching tubes are partially overlapped, and the overlapped part will result in two zero vectors V₀、V₇Meanwhile, the fault, the duty ratio cannot be adjusted, and the reference voltage rotating vector V cannot be recovered_ref ^*Fault-tolerant control cannot be performed in these sectors. For the sector affected by fault switch tube and having at least one undistorted zero vector, making software fault-tolerant control, redefining fault sector T_a，T_b，T_cTime variable, compensation of fault vector in each fault sector, i.e. replacement of fault vector based on specific equivalence principle by using effective vector, rotating reference voltage by vector V_ref ^*It can be resynthesized or approximately recovered in the failed sector. Two upper tubes and one lower tube of different bridge arms simultaneously fail, and one upper tube and two lower tubes of different bridge arms simultaneously fail. Because all sectors are affected by the fault switch tube, normal sectors do not exist, and fault-tolerant buffer control cannot be carried out.

Example 5

The embodiment adopts a three-phase two-level PWM rectifierFour-pipe failures or above are illustrated in the twelve sector partitioning approach. When four or more switching tubes in the three-phase two-level PWM rectifier have faults at the same time, as shown in FIG. 2, the area of the sector overlapped by the faulty switching tubes reaches more than half of the total sector area, and the zero vector V is zero₀、V₇And meanwhile, fault-tolerant control cannot be carried out on the failed sector, and the system is stopped to operate and the switching tube is replaced.

Claims

1. The sector buffer-based two-level PWM rectifier fault-tolerant control method is characterized by comprising the following steps of:

step 1, selecting a sector division mode to carry out sector division;

the sector division mode of the step 1 comprises eight sectors and twelve sectors, and the specific process of the two sectors division is as follows:

(1) adopts an eight-sector division mode

Six variables are defined:

in the formula of U_α、U_βVoltage components of α and β in a two-phase stationary coordinate system;

defining a sign function:

wherein i ═ a, B, C, D, E, F;

as the faults of the three bridge arms of a, b and c correspond to different sector division coordinate systems, in order to accurately carry out fault-tolerant control on fault sectors, a function N of the three sector division coordinate systems is defined_a,N_b,N_c；

Let N_a＝H(A)+H(B)+4H(C)+3H(D)

Let N_b＝4G(B)+3G(C)+G(D)+H(E)

Let N_c＝3H(B)+H(C)+4H(D)+H(F)

TABLE 1 calculation of value N_a,N_b,N_cCorresponding relation with sector

Sector numbering I Ⅱ Ⅲ Ⅳ V Ⅳ Ⅶ Ⅷ Calculating the value N_a 6 2 4 3 7 5 1 8 Calculating the value N_b 2 4 3 7 5 6 1 8 Calculating the value N_c 4 3 7 5 6 2 1 8

selecting different sector division functions according to the position of a fault switch tube, and switching as an a-phase bridge armSelecting a sectorization function N for pipe failures_aThrough N_aDetermining the change sequence of the actual sector numbers according to the change sequence to obtain the division of eight sectors; when a b-phase bridge arm switching tube fault selects a sector division function N_bThrough N_bDetermining the change sequence of the actual sector numbers according to the change sequence to obtain the division of eight sectors; when a c-phase bridge arm switching tube fault selects a sector division function N_cThrough N_cDetermining the change sequence of the actual sector numbers according to the change sequence to obtain the division of eight sectors;

(2) adopts a twelve-sector division mode

Six variables are defined:

defining a sign function:

wherein i ═ a, B, C, D, E, F;

let N ═ sign (A) + sign (B) +2sign (C) +2sign (D) +4sign (E) +3sign (F)

TABLE 2 calculated value N and sector corresponding relation

Calculating the value N 1 2 3 4 5 6 7 8 9 10 11 12 Sector numbering Ⅷ Ⅱ Ⅲ Ⅶ Ⅳ Ⅸ Ⅻ I Ⅹ Ⅵ V Ⅺ

Determining a rotating reference vector V_ref ^*When rotating the reference vector V_ref ^*After one rotation, the change sequence of the calculated values N is as follows: 8 → 4 → 2 → 1 → 3 → 6 → 5 → 9 → 11 → 12 → 10 → 7 → 8, i.e. the order of change of the actual sector numbers, i.e. the division of the twelve sectors;

(1) adopts an eight-sector division mode

(2) adopts a twelve-sector division mode

for the sectors which are affected by a plurality of fault voltage vectors and have no simultaneous fault in the zero vector, the normal zero vector is used for replacing the fault zero vector, the effective vector without the fault is used, the effective vector action time is calculated based on the compensation principle to synthesize the reference voltage rotation vector V again_ref ^*Realizing the fault-tolerant control of the sector, wherein the compensation principle comprises a mapping method, an equiaxial component method and an equiaxial method;

for the sector which is affected by a plurality of fault vectors and has a fault at the same time, because the sector has no normal zero vector, the output vector can not be adjusted, and the reference voltage rotates the vector V_ref ^*The output module value reaches the maximum, and the sector can not carry out fault-tolerant control;

the specific method for adjusting the action time of the basic voltage vector of the buffer sector comprises the following steps: controlling the vector action time when the fault zero vector is normalized in the buffer sector, replacing the zero vector of the periodic fault with the zero vector which does not have the fault in the buffer sector, simultaneously ensuring that the reference voltage rotating vectors in the buffer sector before and after the change are kept consistent, and calculating the compensation time of the effective vector;

2. The sector buffer based two-level PWM rectifier fault-tolerant control method of claim 1, wherein the method for dividing the sector type in step 2 is as follows: if an eight-sector division mode is adopted, determining the sectors affected by the fault switching tube in the eight sectors, namely fault sectors, according to the tables 3-5; if a twelve-sector division mode is adopted, determining the sector influenced by the fault switch tube in the twelve sectors according to the table 6;

3. The sector buffer-based two-level PWM rectifier fault-tolerant control method of claim 1, wherein step 2 determines the change conditions of basic voltage vectors, namely a fault zero vector and an effective vector, before and after a switching tube fault according to a table 7, and determines a fault voltage vector;

TABLE 7 Voltage vector variation before and after single switch tube failure

4. The sector buffer based two-level PWM rectifier fault tolerant control method of claim 1, wherein step 3 selects 1 or 2 consecutive sectors bordering the failed sector as buffer sectors.

5. The fault-tolerant control method for the two-level PWM rectifier based on sector buffering as claimed in claim 1, wherein the specific method for determining the action time of the basic voltage vector of each sector before the fault in the step 4 is

First, the intermediate variables are defined as:

in the formula (I), the compound is shown in the specification,

TABLE 11 sector vs. base Voltage vector action time relationship

6. the sector buffer-based two-level PWM rectifier fault-tolerant method of claim 1, wherein the method for determining the conduction time of the switching tube of each sector in step 6 is as follows:

(1) adopts an eight-sector division mode

for zero vector affected by effective vectorWithout simultaneous failure sectors, it is only necessary to change the conduction time of the three-phase switching tube in the sectors affected by the zero vector, redefining T according to the action time calculated by the compensation principle_a,T_b,T_cDifferent definition formulas of the compensation principle are different;

(a) the mapping method is a method commonly used in the prior art, and how to redefine T is not described herein again_a,T_b,T_c；

redefining T for buffer sectors_a,T_b,T_cWhen the position of the fault switch tube is the upper bridge arm, redefining as:

then, determining the switching time of each sector according to tables 12-14;

(2) Adopts a twelve-sector division mode

for the sector which is affected by a plurality of fault voltage vectors and has no simultaneous fault of the zero vector, the conduction time of the three-phase switch tube is changed only in the sector affected by the zero vector, and the T is redefined according to the action time calculated by the compensation principle_a,T_b,T_cDifferent definition formulas of the compensation principle are different, and the compensation principle is specifically the same as the calculation formula of the eight-sector type sector;

then, determining the switching time of each sector according to the table 15;

TABLE 15 on-off time distribution relationship of different sectors of twelve sectors

7. The sector buffer-based two-level PWM rectifier fault-tolerant control method according to claim 1, characterized in that, in step 7, isosceles triangle waves with switching tube turn-on time and sampling period as sampling periods are modulated, a DWPM technology is adopted, and a vector action sequence is determined based on a symmetry principle and a THD minimum principle, so that 6 paths of PWM pulses can be obtained, the output 6 paths of PWM pulses are acted on a power switching tube driving circuit, and the driving circuit drives corresponding power switching tubes to be turned on and off, so that buffer fault-tolerant control is completed.