CN104092398B - For the SVPWM space vector searching method that three-level current transformer controls - Google Patents

Abstract

The invention discloses a kind of SVPWM space vector searching method controlled for three-level current transformer in current transformer pulse-width controlled technical field. Comprise determining that the subregion that output voltage reference vector is residing in space vector scattergram; Calculate the action time on residing each summit of subregion; Determine the vector state on each summit of residing subregion; Determine the vector sequence of operation of residing subregion and the action time of each vector; According to the action time of described vector sequence of operation and vector, the switch of three-level current transformer is controlled. The present invention utilizes specific storage organization, it is achieved that the purpose of space vector fast search, and the solution procedure of the action time of each vector is not had cyclic search or division arithmetic, greatly speeds up calculating process.

Description

For the SVPWM space vector searching method that three-level current transformer controls

Technical field

The invention belongs to current transformer pulse-width controlled technical field, particularly relate to a kind of SVPWM space vector searching method controlled for three-level current transformer.

Background technology

Along with a large amount of nonlinear-loads put into operation of power networks, as: the illumination of Switching Power Supply, large scale business, Alternating Current Governor System, induction heating equipment etc., these are non-linear, the input of the load of impact causes power network current wave distortion, cause a series of power quality problem of voltage fluctuation and flicker and three-phase imbalance etc. The harm that grid equipment is caused by harmonic wave has: 1) increases line loss and even jeopardizes line security operation; 2) protective relaying device malfunction or tripping are caused; 3) speed-up condenser is aging; 4) accuracy of metering, reasonability are affected; 5) interference communication system, affects communication quality etc. Carrying out the reasonable scheme of power quality controlling in load side is install current transformer additional. Current transformer can detect the harmonic current being injected electrical network by load in real time, the electric current identical, in opposite direction with this harmonic component size is exported by the inverter bridge inversion of self, to offset the harmonic wave that load sends, electrical network only need to provide the fundametal compoment of load, reaches to filter the purpose of harmonic wave.

Current transformer many employings pulse-width controlled technology (PulsewithModulation, PWM) as inversion controlling method, the PWM control methods of Multilevel Inverters mainly has three classes: carrier modulation method, elimination particular harmonic method and space voltage vector PWM control methods (SpaceVectorPWM, SVPWM). SVPWM is with its motility on current transformer DC side derided capacitors mid-point voltage controls and is prone to the advantages such as Digital Implementation, is widely applied in three-level current transformer.

SVPWM is the angle from motor, is conceived to the circular magnetic field how making motor obtain constant amplitude, i.e. sinusoidal magnetic flux. When it is with three-phase symmetrical sinusoidal fluctuation power voltage supply, the desirable magnetic flux circle of ac motor is for benchmark, goes to approach basic circle magnetic flux with the produced actual magnetic flux of the switching mode that current transformer is different, they result of the comparisons determine the switch of current transformer, forms PWM waveform.

In three-phase symmetrical system, three-phase voltage can represent with following formula:

$\left\{\begin{array}{c}{u}_A=U_m\sin \left(\mathrm{\ωt}\right)\\ u_B=U_m\sin (\mathrm{\ωt}-\frac{2}{3}\mathrm{\π})\\ u_C=U_m\sin (\mathrm{\ωt}+\frac{2}{3}\mathrm{\π})\end{array}\right.---\left(1\right)$

In formula (1), u_A、u_BAnd u_CRespectively three-phase voltage, U_mFor the maximum of three-phase voltage amplitude, ω is angular velocity and ω=2 π f, f are line voltage rated frequency.

Corresponding space voltage vector is defined as:

$\stackrel{\·}{U}=\frac{2}{3}(u_A+\mathrm{\α}{u}_B+{\mathrm{\α}}^2{u}_C)---\left(2\right)$

In formula (2),J is imaginary unit.

The switch models of desirable three-level current transformer circuit becomes a SP3T switch S communicated with direct current as it is shown in figure 1, the circuit structure of every phase brachium pontis can simplify. In Fig. 1, S_a、S_bAnd S_cRepresent the output state of each brachium pontis.

As shown in Figure 1, each phase voltage is expressed as:

$\left\{\begin{array}{c}{u}_A=\frac{E}{2}{S}_a\\ u_B=\frac{E}{2}{S}_b\\ u_C=\frac{E}{2}{S}_c\end{array}\right.---\left(3\right)$

In formula (3), E is the voltage of dc bus, and has:

In formula (4), x is a, b and c three-phase.

By above formula it can be seen that three-phase tri-level current transformer can export 27 kinds of voltage status combinations, corresponding converter switches state 27 groups different. Still defining space vector of voltage is:

$\stackrel{\·}{U}\left(k\right)=\frac{1}{3}E(S_a+\mathrm{\α}{S}_b+{\mathrm{\α}}^2{S}_c)=\frac{E}{6}[(2S_a-S_b-S_c)+j\sqrt{3}(S_b-S_c)]---\left(5\right)$

Above formula is by the formula of voltage vector discretization, and k is the voltage vector of kth sampled point. In alpha-beta plane, the space vector scattergram corresponding to 27 groups of on off states is as shown in Figure 2.

In Fig. 2, space vector is divided into six big sectors of A, B, C, D, E and F equalization, each big sector is divided into again 4 impartial little sectors (hereinafter referred to as " subregion "), and according to from the inside to the outside, counter clockwise direction determines the order of the subregion in each big sector. It is divided into tetra-subregions of I, II, III, IV for big sector A, A. In the three-level current transformer alpha-beta space vector plane shown in Fig. 2, by the difference of basic vector, 27 groups of on off states of 19 basic vectors and correspondence thereof are divided into four classes, respectively long vector, middle vector, short vector zero vector, as shown in Fig. 3 table provided.

In order to make the voltage vector that three-level current transformer exports be close to round, and finally give the rotating magnetic flux chain of circle, space vector (output-voltage levels) and the combination of action time of the current transformer obtained can be utilized, go close to circular with polygon. Traditional SVPWM adopts the method for similar two level to realize the SVPWM of three level, it is necessary to relate to more trigonometric function operation, requires higher to the arithmetic speed controlling system.

Tradition SVPWM has been improved by 60 ° of coordinate system algorithms, adopts non-orthogonal 60 ° of coordinate systems can simplify synthesis and the calculating of action time of reference vector. For convenience, 60 ° of coordinate systems being denoted as g-h coordinate system, take g axle and α axle overlaps in alpha-beta plane, rotating 60 ° counterclockwise is h axle, as shown in Figure 4.

For three-phase output voltage reference vector, under ABC coordinate system, three-phase voltage is expressed as u_A, u_B, u_C, the voltage vector form obtained in g-h coordinate system through clark conversion is:

$\left[\begin{array}{c}{u}_g\\ u_h\end{array}\right]=\sqrt{\frac{2}{3}}\left[\begin{array}{ccc}1& -1& 0\\ 0& 1& -1\end{array}\right]\left[\begin{array}{c}{u}_A\\ u_B\\ u_C\end{array}\right]---\left(6\right)$

By above formula it can be seen that representation space voltage vector can be carried out with integral point coordinate under g-h coordinate system. Assume that DC voltage is u_d, willDivided byIt is normalized so thatNormalizing value drop within an equilateral hexagon, hexagon maximum isNow, center is 2 to the length on any one summit. The coordinate on each summit, each sector under g-h coordinate system and corresponding voltage vector is obtained, as shown in Figure 5 after above-mentioned process.

As shown in Figure 5, the closer to initial point, the redundancy of space vector is more many, formula (5) know, it is necessary to solving equations just can determine that space vector. Additionally, need the conversion of (floating-point and integer) between data type when asking for each vector action time, compare and the computing such as judgement, in the occasion harsher to calculating rate request, the application of 60 ° of coordinate system algorithms is subject to certain restrictions.

Summary of the invention

It is an object of the invention to, a kind of SVPWM space vector searching method controlled for three-level current transformer is provided, a large amount of cyclic searches operation when avoiding existing SVPWM to solving equations, eliminates the checking procedure to current transformer output vector, alleviates the calculating pressure of controller.

To achieve these goals, the technical scheme that the present invention proposes is, a kind of SVPWM space vector searching method controlled for three-level current transformer is characterized in that described method includes:

Step 1: determine the subregion that output voltage reference vector is residing in space vector scattergram;

Step 2: calculate the action time on residing each summit of subregion;

Step 3: determine the vector state on each summit of residing subregion;

Described vector state includes vector value and vector number;

Step 4: determine the vector sequence of operation of residing subregion and the action time of each vector;

Step 5: the switch of three-level current transformer is controlled according to the action time of described vector sequence of operation and vector.

Action time on each summit of subregion residing for described calculating particularly as follows:

When the subregion I that subregion is sector A residing in space vector scattergram of output voltage reference vector, the subregion II of sector A, sector A the subregion III of subregion IV or sector F time, the action time on residing each summit of subregion is respectively as follows: $\left\{\begin{array}{c}{d}_1=g_r-h_3\\ d_2=h_r-g_3\\ d_3=1-(d_1+d_2)\end{array}\right.;$

When the subregion I that subregion is sector B residing in space vector scattergram of output voltage reference vector, the subregion II of sector B, sector B the subregion III of subregion IV or sector E time, the action time on residing each summit of subregion is respectively as follows: $\left\{\begin{array}{c}{d}_1={-(h}_r-h_2)\\ d_2=-(g_r-g_2)\\ d_3=1-(d_1+d_2)\end{array}\right.;$

When the subregion I that subregion is sector C residing in space vector scattergram of output voltage reference vector, the subregion II of sector C, sector C the subregion III of subregion IV or sector D time, the action time on residing each summit of subregion is respectively as follows: $\left\{\begin{array}{c}{d}_1=g_r-h_1\\ d_2=h_r-g_1\\ d_3=1-(d_1+d_2)\end{array}\right.;$

When the subregion I that subregion is sector D residing in space vector scattergram of output voltage reference vector, the subregion II of sector D, sector D the subregion III of subregion IV or sector C time, the action time on residing each summit of subregion is respectively as follows: $\left\{\begin{array}{c}{d}_1={-(h}_r-h_3)\\ d_2={-(g}_r-g_3)\\ d_3=1-(d_1+d_2)\end{array}\right.;$

When the subregion I that subregion is sector E residing in space vector scattergram of output voltage reference vector, the subregion II of sector E, sector E the subregion III of subregion IV or sector B time, the action time on residing each summit of subregion is respectively as follows: $\left\{\begin{array}{c}{d}_1=g_r-h_2\\ d_2=h_r-g_2\\ d_3=1-(d_1+d_2)\end{array}\right.;$

When the subregion I that subregion is sector F residing in space vector scattergram of output voltage reference vector, the subregion II of sector F, sector F the subregion III of subregion IV or sector A time, the action time on residing each summit of subregion is respectively as follows: $\left\{\begin{array}{c}{d}_1={-(h}_r-h_1)\\ d_2=-(g_r-g_1)\\ d_3=1-(d_1+d_2)\end{array}\right.;$

Wherein, d₁、d₂And d₃Respectively residing subregion summit p₁、p₂And p₃Action time, g₁And h₁Respectively summit p₁G coordinate and h coordinate, g₂And h₂Respectively summit p₂G coordinate and h coordinate, g₃And h₃Respectively summit p₃G coordinate and h coordinate, g_rAnd h_rThe respectively g coordinate of output voltage reference vector and h coordinate.

The described vector sequence of operation determining residing subregion specifically, from subregion corresponding with vector sequence of operation storage organization is searched residing subregion press the vector identification that vector sequence of operation sorts; Further according to vector identification, vector identification storage organization is searched the vector that vector identification is corresponding.

Described subregion and vector sequence of operation storage organization are array, chained list, class or tables of data.

Described vector identification storage organization is array, chained list, class or tables of data.

The process of setting up of described vector identification storage organization is:

Step A1: set up the first storage organization, stores in the first storage organization by the sector mark belonging to each subregion, partition identification and each summit of the subregion coordinate in g-h coordinate system;

Step A2: by g-h coordinate system to two units of right translation, obtain g '-h ' coordinate system;

Step A3: set up the second storage organization, stores in the second storage organization by the sector mark belonging to each subregion, partition identification and each summit of the subregion coordinate in g '-h ' coordinate system;

Step A4: set up the 3rd storage organization, stores in the 3rd storage organization by the vector state on the summit of each subregion g ' coordinate in g '-h ' coordinate system, h ' coordinate and summit;

Step A5: set up vector identification storage organization, the vector on each each summit of subregion is extracted from the 3rd storage organization, and be that the vector of each extraction sets vector identification, then vector identification corresponding for vector described in the vector on each each summit of subregion is stored in vector identification storage organization.

Described first storage organization, the second storage organization and the 3rd storage organization are array, chained list, class or tables of data.

The process of setting up of described subregion and vector sequence of operation storage organization is:

Step B1: determine the vector sequence of operation of each subregion, and according to described vector sequence of operation, vector identification is sorted;

Step B2: set up subregion and vector sequence of operation storage organization, stores the sector mark belonging to each subregion, partition identification and the vector identification sorted by vector sequence of operation in subregion and vector sequence of operation storage organization.

The present invention utilizes specific storage organization, it is achieved that the purpose of space vector fast search, and the solution procedure of the action time of each vector is not had cyclic search or division arithmetic, greatly speeds up calculating process.

Accompanying drawing explanation

Fig. 1 is three-level current transformer switch models circuit diagram;

Fig. 2 is three-level current transformer space voltage vector scattergram;

Fig. 3 is the vector classification table of three-level current transformer;

Fig. 4 is g-h coordinate system and alpha-beta ordinate transform graph of a relation;

Fig. 5 is coordinate and the 3 level space vector figure of g-h coordinate system;

Fig. 6 is space vector each subregion apex coordinate storage order table under g-h coordinate system;

Fig. 7 is the 3 level space vector figure under the g '-h ' coordinate system obtained after g-h coordinate system 2 units of translation;

Fig. 8 is the coordinate storage order table on each subregion summit of space vector under g '-h ' coordinate system;

Fig. 9 is space vector and the storage order table thereof on each summit under g '-h ' coordinate system;

Figure 10 is the data structure schematic diagram that output reference voltage vector is in 7 segmentation vectors of subregion AIII;

The region that Figure 11 is space voltage vector divides schematic diagram;

Figure 12 is output voltage vector mark and the storage order table thereof of each subregion;

Figure 13 is the SVPWM space vector searching method flow chart controlled for three-level current transformer;

Figure 14 is the data structure schematic diagram that output reference voltage vector is in 7 segmentation vectors of subregion AII;

Figure 15 is the inverter bridge circuit structure diagram of three-phase tri-level current transformer.

Detailed description of the invention

Below in conjunction with accompanying drawing, preferred embodiment is elaborated. It is emphasized that the description below is merely exemplary, rather than in order to limit the scope of the present invention and application thereof.

The enforcement of the present invention divides two parts, and one is the design of storage organization, and two is the fast search process of space vector.

First, space vector being divided into six big sectors of equalization under g-h coordinate system, each big sector of space vector is an equilateral triangle, and the mark of these equilateral triangles is respectively set as A, B, C, D, E and F.

It follows that more each big sector is divided into 4 impartial subregions, each subregion is again a little equilateral triangle.The partition identification comprising midpoint (zero vector point) in each sector is set as I, the flag of the subregion being positioned at each sector central authorities is III, other partition identification in each sector are respectively set as II and IV, and mark is all II or mark be all two subregions of IV can not be adjacent.

The space vector that g-h coordinate system divides is as shown in Figure 5. Afterwards, it is the design process of concrete storage organization. The present embodiment will using array as data store organisation, and what illustrate vector identification storage organization and subregion and vector sequence of operation storage organization sets up process.

The process of setting up of vector identification storage organization includes:

Step A1: set up the first array (OrigCoordTable), the sector mark belonging to each subregion of memory space vector, partition identification and each summit of the subregion coordinate in g-h coordinate system.

First array is the three-dimensional array of 6 × 4 × 6, and its first Wesy is in the sector mark belonging to partition holding, and the second Wesy identifies in partition holding, and the third dimension is used for partition holding each summit coordinate in g-h coordinate system.

Owing to each subregion has three summits, therefore the order of three summit storages is as follows:

(1) storage order on three summits of subregion I is:

Centered by space vector midpoint (zero vector), make straight line, with midpoint for the center of circle, rotate this straight line by clockwise direction, straight line is labeled as p through first summit (non-midpoint) of subregion I₁(g₁, h₁), it is labeled as p through second summit (non-midpoint) of subregion I₂(g₂, h₂), midpoint is labeled as p₃(g₃, h₃). According to p₁、p₂And p₃Sequential storage each point g-h coordinate.

Such as, in Fig. 5, midpoint (0 is crossed as straight line, 0) and point (-1,2), this straight line is rotated by clockwise direction, three summits for the subregion I (under be called for short AI) of sector A, straight line is (0,1) through first summit (non-midpoint) of AI, is designated as p₁, straight line is (1,0) through second summit (non-midpoint) of AI, is designated as p₂, midpoint (0,0) is designated as p₃, by p₁、p₂And p₃The above-mentioned g-h coordinate of sequential storage be (0,1,1,0,0,0).

(2) storage order on three summits of subregion II or IV is:

Centered by space vector midpoint (zero vector), make straight line, with midpoint for the center of circle, rotate this straight line by clockwise direction.

If straight line is through a summit (non-midpoint) of subregion II or IV, then it is marked as p₁, straight line continues this straight line that turns clockwise, if (now marked p through two points simultaneously₁), then by two points of straight line process simultaneously, the apex marker that h coordinate absolute value is big is p₂, the apex marker that h coordinate absolute value is little is p₃. If in two points of straight line process simultaneously, h coordinate figure is equal, then it is p by apex marker big for g coordinate absolute value₂, the apex marker that g coordinate absolute value is little is p₃。

If straight line is simultaneously through two points (now unmarked p₁), then by straight line simultaneously in two points, the apex marker that h coordinate absolute value is big is p₁, the apex marker that h coordinate absolute value is little is p₃If in two points of straight line process simultaneously, h coordinate figure is equal, then it is p by apex marker big for g coordinate absolute value₁, the apex marker that g coordinate absolute value is little is p₃. Straight line continues to rotate by clockwise direction, and the next point through subregion II or IV is labeled as p₂。

Such as, in Fig. 5, midpoint (0 is crossed as straight line, 0) and point (-1,2), this straight line is rotated by clockwise direction, three summits for the subregion II (under be called for short AII) of sector A, straight line is (1,1) through first summit (non-midpoint) of AII, is designated as p₁.Straight line continues to rotate clockwise, simultaneously through point (1,0) and point (2,0), due to now labeled p₁, and the h coordinate of point (1,0) and point (2,0) is equal, therefore summit (2,0) big for the absolute value of point (1,0) and the g coordinate of point (2,0) is labeled as p₂, the summit (1,0) that the absolute value of g coordinate is little is labeled as p₃. By p₁、p₂And p₃The above-mentioned g-h coordinate of sequential storage be (1,1,2,0,1,0).

Again such as, in Fig. 5, made straight line is with upper identical and rotate this straight line by clockwise direction, three summits for the subregion II (under be called for short FII) of sector F, straight line is (2 ,-1) through first summit (non-midpoint) of FII, is designated as p₁. Straight line continues to rotate clockwise, simultaneously through point (2 ,-2) and point (1 ,-1), due to now labeled p₁, therefore summit (2 ,-2) big for the absolute value of point (2 ,-2) and the h coordinate of point (1 ,-1) is labeled as p₂, the summit (1 ,-1) that the absolute value of h coordinate is little is labeled as p₃. By p₁、p₂And p₃The above-mentioned g-h coordinate of sequential storage be (2 ,-1,2 ,-2,1 ,-1).

Again such as, in Fig. 5, made straight line is with upper identical and rotate this straight line by clockwise direction, for three summits of the subregion IV (under be called for short AIV) of sector A, straight line is simultaneously through the point (0,2) of AIV and point (0,1), p is crossed due to now unmarked₁, therefore summit (0,2) big for the absolute value of point (0,2) and the h coordinate of point (0,1) is labeled as p₁, the summit (0,1) that the absolute value of h coordinate is little is labeled as p₃. Continuing the straight line that turns clockwise, straight line, through point (1,1), is marked as p₂. By p₁、p₂And p₃The above-mentioned g-h coordinate of sequential storage be (0,2,1,1,0,1).

Again such as, in Fig. 5, made straight line is with upper identical and rotate this straight line by clockwise direction, for three summits of the subregion IV (under be called for short FIV) of sector F, straight line is simultaneously through the point (2,0) of FIV and point (1,0), p is crossed due to now unmarked₁, and the h coordinate figure of point (2,0) and point (1,0) is equal, therefore summit (2,0) big for the absolute value of point (2,0) and the g coordinate of point (1,0) is labeled as p₁, the summit (1,0) that the absolute value of g coordinate is little is labeled as p₃. Continuing the straight line that turns clockwise, straight line, through point (2 ,-1), is marked as p₂. By p₁、p₂And p₃The above-mentioned g-h coordinate of sequential storage be (2,0,2 ,-1,1,0).

(3) storage order on three summits of subregion III is:

The coordinate of the middle vector (middle vector provides in the table of Fig. 3) in three dimensional vector diagram is labeled as p₁(g₁, h₁), at p₁And between zero vector, carry out line, with zero vector for axle, first apex marker of the subregion III that line rotates past counterclockwise is p₂(g₂, h₂), first apex marker of the subregion III that line rotates clockwise past is p₃(g₃, h₃). By p₁、p₂And p₃The above-mentioned g-h coordinate of sequential storage.

Such as, in Fig. 5, in three summits of the subregion III (under be called for short AIII) of sector A, middle vector is (1,1), is marked as p₁, at p₁And between zero vector, carry out line, with zero vector for axle, first summit (0,1) of the AIII that line rotates past counterclockwise is labeled as p₂, the summit (1,0) of the AIII that line rotates clockwise past is labeled as p₃. By p₁、p₂And p₃The above-mentioned g-h coordinate of sequential storage be (1,1,0,1,1,0).

Through the process of above (1st), (2nd) and (3rd) part, the summit p of each subregion that will obtain₁、p₂And p₃Coordinate insert in OrigCoordTable, form is ((A～F), (I～IV), (g₁, h₁, g₂, h₂, g₃, h₃)), wherein A～F represents in sector mark A～F, and I～IV represents in partition identification I～IV, lower same.Shown in the table that the data finally arranged such as Fig. 6 provides.

Step A2: by g-h coordinate system to two units of right translation, obtain g '-h ' coordinate system.

In Figure 5, g axle, h axle coordinate range be-2～2, in most computer programming language, the subscript of array is all 0 beginning, active computer carry out vector preservation and search before also need to do a few thing. The technical scheme is that the coordinate in Fig. 5 to 2 units of right translation, namely the apex coordinate corresponding to each sector in g-h coordinate carries out adding 2 computings, obtains new coordinate g '-h ', as shown in Figure 7, in new coordinate system, coordinate range becomes 0～4, it is simple to computer disposal.

Step A3: set up the second array (TranCoordTable), for storing sector mark belonging to each subregion, partition identification and each summit of the subregion coordinate in g '-h ' coordinate system.

Second array is also the three-dimensional array of 6 × 4 × 6, and the first Wesy of the second array is in the sector mark belonging to partition holding, and the second Wesy identifies in partition holding, and the third dimension is used for partition holding each summit coordinate in g '-h ' coordinate system. It practice, the data of the third dimension storage of the second array are that the data that the third dimension to the first array stores have carried out adding 2 process. Shown in the table that TranCoordTable after arrangement such as Fig. 8 provides.

Step A4: set up the 3rd array (VectorTable), is used for the vector state on the summit storing each subregion g ' coordinate in g '-h ' coordinate system, h ' coordinate and summit. Wherein, the vector state on summit refers to the vector value on summit and the vector number on summit.

3rd array is the three-dimensional array of 5 × 5 × 7, and its first Wesy is in the g ' coordinate figure on storage summit, and the second Wesy is in the h ' coordinate figure on storage summit, and the third dimension is for storing the vector state on summit. The third dimension includes 7 digit order numbers, it is possible to store 7 numerals. If the number of vector is 1 on summit, then front 3 digit order numbers of the third dimension store this vector value, the 4-6 digit order number storage 0 value; If the number of vector is 2 on summit, then front 3 digit order numbers of the third dimension store the vector value of the 1st vector, and the 4-6 digit order number stores the vector value of the 2nd vector; If the number of vector is 0 on summit, or only have zero vector on summit, then the 1-6 digit order number all stores 0 value. On 7th digit order number storage summit, the number of vector, when the 7th digit order number is 0 value, illustrates that this summit is not in voltage vector-diagram.

Such as, in Fig. 7, the summit (2,4) of AIV, its g ' coordinate is 2, and h ' coordinate is 4, and vector thereon only has 1, and vector is ppn, and numeral represents that vector value is 1,1 ,-1. Then the vector state of summit (2,4) is (1,1 ,-1,0,0,0,1).

Again such as, in Fig. 7, the summit (3,2) of AIII, its g ' coordinate is 3, and h ' coordinate is 2, and vector thereon has 2, and vector is poo and onn respectively, and numeral represents vector value respectively 1,0,0 and 0 ,-1 ,-1. Then the vector state on the summit of summit (3,2) is (1,0,0,0 ,-1 ,-1,2).

In the manner described above, the 3rd array of data is stored as shown in Figure 9.

Step A5: set up the 4th array (SerialVector), for storing the vector vector correspondence vector identification on each each summit of subregion.

4th array is two-dimensional array, and the first Wesy is in storage vector identification, and the second Wesy is in storage vector. According to three summits of each subregion coordinate in g '-h ' coordinate system, searching the corresponding vector state in summit in the 3rd array, after obtaining vector value, storage is to the 4th array relevant position.

If the corresponding vector number in summit found in the 3rd array is 1, then the corresponding vector value in this summit leaves in front 3 digit order numbers of the 3rd array third dimension data, is extracted. If vector identification 0 did not use, then and by this vector identification it is set to 0; If mark 0 already with, then identify as 1.

If the corresponding vector number in summit found in the 3rd array is 2, then the corresponding vector value in this summit leaves in front 3 digit order numbers and the 4-6 digit order number of the 3rd array third dimension data respectively, they is extracted. If vector identification 2 and 3 did not use, then the vector identification of the two vector is set as 2 and 3; If vector identification 2 and 3 is already with mistake, then the vector identification of the two vector is set as 4 and 5.

If the vector number of the vertex correspondence found in the 3rd array is 3, then the corresponding vector value in this summit is 0, and this vector is set as ppp, and vector identification is set as 6.

Afterwards, vector identification corresponding for this vector of vector is stored in the 4th array.

For the ii I subregion AIII of sector A, in g '-h ' coordinate system, the apex coordinate of AIII respectively (3,3), (2,3) and (3,2), the 3rd array is searched the corresponding vector state in summit. Finding the corresponding vector state in summit (3,3) is (1,0 ,-1,0,0,0,1), the corresponding vector state in summit (2,3) is (1,1,0,0,0 ,-1,2), the corresponding vector state in summit (3,2) is (1,0,0,0 ,-1 ,-1,2). Next according to above-mentioned storage principle, vector is stored in the 4th array. Owing to (1,0 ,-1,0,0,0,1) last value is 1, namely vector number is 1, and vector identification 0 does not use, and therefore extracts its first three digit order number 1,0 ,-1, and its vector is pon, and vector identification is 0. Due to (1,1,0,0,0 ,-1,2) last value is 2, and namely vector number is 2, and vector identification 2 and 3 does not use, and therefore extracts first three digit order number 1,1 respectively, 0 and the 4-6 digit order number 0,0 ,-1, its vector respectively ppo and oon, vector identification respectively 2 and 3. Due to (1,0,0,0 ,-1 ,-1,2) last value is 2, and namely vector number is 2, and vector identification 2 and 3 already with, therefore extract first three digit order number 1,0 respectively, 0 and the 4-6 digit order number 0 ,-1 ,-1, its vector respectively poo and onn, vector identification respectively 4 and 5. Above-mentioned vector vector identification is stored in after the 4th array as shown in Figure 10.

The process setting up subregion and vector sequence of operation storage organization is:

Step B1: determine the vector sequence of operation of each subregion, and according to described vector sequence of operation, vector identification is sorted.

Each sequence of operation organizing vector should follow a principle: the change of any voltage vector can only have the switch motion of a brachium pontis, in binary vector represents, during on off state switching in office, and the voltage level change of only one of which brachium pontis. If this is because allow the action simultaneously of two or more brachium pontis, then there will be the pulse of reversed polarity in the half period of on-Line Voltage, the opposing torque of this pulses generation will cause pulsation and electromagnetic noise. 3 level space vector figure is divided into outer shroud, medium ring and internal ring by technical scheme, and as shown in figure 11, dash area is medium ring, and internal ring, in the inside of medium ring, is an equilateral hexagon, and remaining part is outer shroud. Select the vector on internal ring summit as originating vector in each sampling period, adopt the rising edge pulse of seven segmentations, it is the same that end vector originates vector, therefore, owing to current transformer sample rate is all higher, when output reference voltage moves between any subregion in adjacent two big sectors, the switch motion at the maximum only one of which brachium pontis of synchronization of each brachium pontis can be ensured.

The present embodiment selects the vector ppo on the summit (0,1) in g-h coordinate system as starting vector, certainly, selects the arbitrary vector in oon, poo, onn to be all possible, but must meet three conditions:

1) vector on three summits will be used.

2) neighbouring vectors only has the switch motion of a phase, such as ppo → poo, the only switch motion of B phase.

3) in order to enable export in dsp for the IGBT pulse driven, vector sequence must be symmetrical, in vector sequence ppo → poo → pon → oon → pon → poo → ppo, symmetrical centered by oon.

The present embodiment selects short vector as starting vector, i.e. summit (1,0), (0,1), (-1,1), (-1,0), (0,-1), the vector on (1 ,-1), having two vectors to use on these summits, such benefit is:

1) convenience that the IGBT driving pulse of output symmetry and DSP realize.

2) reduce the IGBT number of times cut-off, reduce loss.

3) two vectors on each summit, output action effect is the same, but the action effect of dc-link capacitance is completely contrary, their action time can as the variable of dc-link capacitance mid-point voltage control.

According to mentioned above principle, for the ii I subregion AIII of sector A, correct vector sequence of operation should be ppo → poo → pon → oon → pon → poo → ppo. According to this vector sequence of operation, vector identification is sorted. In the 4th array, finding the vector on each summit of subregion AIII and the vector identification (as shown in Figure 10) of correspondence thereof, according to above-mentioned vector sequence of operation, vector identification is sorted, the result obtained is: 2,4,0,3,0,4,2.

Step B2: set up the 5th array, stores in the 5th array by the sector mark belonging to each subregion, partition identification and the vector identification sorted by vector sequence of operation, and the 5th array after storage is as shown in figure 12.

Finally, the 5th array set up is utilized, it is achieved the search of SVPWM space vector. With output voltage reference vectorResiding sector is A and subregion is II, i.e. output voltage reference vectorBeing positioned at subregion AII is example, and the search procedure of space vector is described. As shown in figure 13, this process includes the search routine figure of space vector:

Step 1: determine the subregion that output voltage reference vector is residing in space vector scattergram.

According to it is assumed that output voltage reference vectorResiding subregion is AII, ifCoordinate under g-h coordinate is (g_r,h_r)。

Step 2: calculate the action time on residing each summit of subregion.

The calculation of the action time on each summit of subregion is as follows:

When the subregion I that subregion is sector A residing in space vector scattergram of output voltage reference vector, the subregion II of sector A, sector A the subregion III of subregion IV or sector F time, the action time on residing each summit of subregion is respectively as follows:

$\left\{\begin{array}{c}{d}_1=g_r-h_3\\ d_2=h_r-g_3\\ d_3=1-(d_1+d_2)\end{array}\right.---\left(7\right)$

When the subregion I that subregion is sector B residing in space vector scattergram of output voltage reference vector, the subregion II of sector B, sector B the subregion III of subregion IV or sector E time, the action time on residing each summit of subregion is respectively as follows:

$\left\{\begin{array}{c}{d}_1={-(h}_r-h_2)\\ d_2=-(g_r-g_2)\\ d_3=1-(d_1+d_2)\end{array}\right.---\left(8\right)$

When the subregion I that subregion is sector C residing in space vector scattergram of output voltage reference vector, the subregion II of sector C, sector C the subregion III of subregion IV or sector D time, the action time on residing each summit of subregion is respectively as follows:

$\left\{\begin{array}{c}{d}_1=g_r-h_1\\ d_2=h_r-g_1\\ d_3=1-(d_1+d_2)\end{array}\right.---\left(9\right)$

When the subregion I that subregion is sector D residing in space vector scattergram of output voltage reference vector, the subregion II of sector D, sector D the subregion III of subregion IV or sector C time, the action time on residing each summit of subregion is respectively as follows:

$\left\{\begin{array}{c}{d}_1={-(h}_r-h_3)\\ d_2={-(g}_r-g_3)\\ d_3=1-(d_1+d_2)\end{array}\right.---\left(10\right)$

When the subregion I that subregion is sector E residing in space vector scattergram of output voltage reference vector, the subregion II of sector E, sector E the subregion III of subregion IV or sector B time, the action time on residing each summit of subregion is respectively as follows:

$\left\{\begin{array}{c}{d}_1=g_r-h_2\\ d_2=h_r-g_2\\ d_3=1-(d_1+d_2)\end{array}\right.---\left(11\right)$

When the subregion I that subregion is sector F residing in space vector scattergram of output voltage reference vector, the subregion II of sector F, sector F the subregion III of subregion IV or sector A time, the action time on residing each summit of subregion is respectively as follows:

$\left\{\begin{array}{c}{d}_1={-(h}_r-h_1)\\ d_2=-(g_r-g_1)\\ d_3=1-(d_1+d_2)\end{array}\right.---\left(12\right)$

Wherein, d₁、d₂And d₃Respectively residing subregion summit p₁、p₂And p₃Action time, g₁And h₁Respectively summit p₁G coordinate and h coordinate, g₂And h₂Respectively summit p₂G coordinate and h coordinate, g₃And h₃Respectively summit p₃G coordinate and h coordinate, g_rAnd h_rThe respectively g coordinate of output voltage reference vector and h coordinate.

Due in the present embodiment, output voltage reference vectorIt is in subregion AII, and the g-h coordinate respectively p on the three of AII summits₁(1,1)、p₂And p (2,0)₃(1,0)。

If vector respectively d action time on three summits₁、d₂And d₃, then the vector on three summits is respectively as follows: action time $\left\{\begin{array}{c}{d}_1=g_r-h_3=g_r-0=g_r\\ d_2=h_r-g_3=h_r-1\\ d_3=1-(d_1+d_2)\end{array}\right..$

Step: 3: determine the vector state on each summit of residing subregion.

Vector state on each summit of subregion AII can directly utilize the 3rd array (VectorTable) and know, as shown in Figure 9. Subregion AII summit (3,3) (p₁Translate gained through 2 units) vector state be (1,0 ,-1,0,0,0,1), subregion AII summit (4,2) (p₂Translate gained through 2 units) vector state be (1 ,-1 ,-1,0,0,0,1), subregion AII summit (3,2) (p₃Translate gained through 2 units) vector state be (1,0,0,0 ,-1 ,-1,2).

The vector identification that the vector vector on above-mentioned each summit of subregion AII is corresponding stores after the 4th array as shown in figure 14. Make V₁For summit p₁On vector, then V₁=pon. V₂For summit p₂On vector, then V₂=pnn. V₃And V₄Respectively summit p₃On vector, V₃=poo and V₄=onn.

Step 4: determine the vector sequence of operation of residing subregion and the action time of each vector.

With short vector V₃=poo sends out vector as originating vector end, and vector sequence must be symmetrical, then the vector sequence of operation of subregion AII is V₃→V₁→V₂→V₄→V₂→V₁→V₃. Above-mentioned vector is respectively as follows: t action time₃、t₁、t₂、t₄、t₂、t₁And t₃, i.e. vector V_iAction time be t_i, i=1,2,3,4. Due to p₁The action time of point is d₁, and vector V₁Act on twice, so vector V₁T action time₁=d₁/ 2. In like manner, vector V₂T action time₂=d₂/2。p₃The action time of point is d₃, due to p₃Point has two vector V₃And V₄Effect, therefore the two vector average distribution p₃D action time of point₃, i.e. vector V₄T action time₄=d₃/ 2. Again due to vector V₃Act on twice, therefore vector V₃Action time be t₃=d₃/4。

Step 5: the switch of three-level current transformer is controlled according to the action time of described vector sequence of operation and vector.

After determining the action time of vector sequence of operation and vector, according to this order and time, the switch of three-level current transformer is controlled, space vector search can be realized.

What the switch of three-level current transformer was controlled is briefly discussed below:

Owing to each vector is made up of 3 letters, the vector of the 1st letter representative acts on A phase, and the vector of the 2nd letter representative acts on B phase, and the 3rd the alphabetical vector represented acts on C phase. As single vector pon, p vector acts on A phase, o vector acts on B phase, and n vector acts on C phase.

In the three-level current transformer main circuit of Figure 15, ABC three-phase symmetrical, for A phase explanation. A phase has 4 IGBT and distinguishes antiparallel diode D with IGBT, and the IGBT A1～A4 of 4 A phases represents, two other diode D plays clamping action.

1) when p vector acts on A phase time, A1, A2 turn on, and A3, A4 turn off, and A phase directly connects with C1 positive pole, export positive level;

2) when o vector acts on A phase time, A2, A3 turn on, and A1, A4 turn off, and A phase is directly connected with the midpoint O of C1 and C2, export zero level;

3) when n vector acts on A phase time, A1, A2 turn off, and A3, A4 turn on, and A phase is directly connected with the negative pole of C2, export negative level.

In the manner described above, according to the action time of vector sequence of operation and vector, the switch of three-level current transformer is controlled, space vector can be obtained.

The above; being only the present invention preferably detailed description of the invention, but protection scope of the present invention is not limited thereto, any those familiar with the art is in the technical scope that the invention discloses; the change that can readily occur in or replacement, all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with scope of the claims.

Claims (7)

1. the SVPWM space vector searching method controlled for three-level current transformer, is characterized in that described method includes:

Step 1: determine the subregion that output voltage reference vector is residing in space vector scattergram;

Step 2: calculate the action time on residing each summit of subregion;

Step 3: determine the vector state on each summit of residing subregion;

Described vector state includes vector value and vector number;

Step 4: determine the vector sequence of operation of residing subregion and the action time of each vector;

Step 5: the switch of three-level current transformer is controlled according to the action time of described vector sequence of operation and vector;

Described step 2 calculates the action time on residing each summit of subregion particularly as follows:

When the subregion I that subregion is sector A residing in space vector scattergram of output voltage reference vector, the subregion II of sector A, sector A the subregion III of subregion IV or sector F time, the action time on residing each summit of subregion is respectively as follows: $\left\{\begin{array}{c}{d}_1=g_r-h_3\\ d_2=h_r-g_3\\ d_3=1-(d_1+d_2)\end{array}\right.;$

When the subregion I that subregion is sector B residing in space vector scattergram of output voltage reference vector, the subregion II of sector B, sector B the subregion III of subregion IV or sector E time, the action time on residing each summit of subregion is respectively as follows: $\left\{\begin{array}{c}{d}_1=-(h_r-h_2)\\ d_2=-(g_r-g_2)\\ d_3=1-(d_1+d_2)\end{array}\right.;$

When the subregion I that subregion is sector C residing in space vector scattergram of output voltage reference vector, the subregion II of sector C, sector C the subregion III of subregion IV or sector D time, the action time on residing each summit of subregion is respectively as follows: $\{\begin{array}{c}{d}_1=g_r-h_1\\ d_2=h_r-g_1\\ d_3=1-(d_1+d_2)\end{array};$

When the subregion I that subregion is sector D residing in space vector scattergram of output voltage reference vector, the subregion II of sector D, sector D the subregion III of subregion IV or sector C time, the action time on residing each summit of subregion is respectively as follows: $\left\{\begin{array}{c}{d}_1=-(h_r-h_3)\\ d_2=-(g_r-g_3)\\ d_3=1-(d_1+d_2)\end{array}\right.;$

When the subregion I that subregion is sector E residing in space vector scattergram of output voltage reference vector, the subregion II of sector E, sector E the subregion III of subregion IV or sector B time, the action time on residing each summit of subregion is respectively as follows: $\left\{\begin{array}{c}{d}_1=g_r-h_2\\ d_2=h_r-g_2\\ d_3=1-(d_1+d_2)\end{array}\right.;$

When the subregion I that subregion is sector F residing in space vector scattergram of output voltage reference vector, the subregion II of sector F, sector F the subregion III of subregion IV or sector A time, the action time on residing each summit of subregion is respectively as follows: $\left\{\begin{array}{c}{d}_1=-(h_r-h_1)\\ d_2=-(g_r-g_1)\\ d_3=1-(d_1+d_2)\end{array}\right.;$

Wherein, d₁、d₂And d₃Respectively residing subregion summit p₁、p₂And p₃Action time, g₁And h₁Respectively summit p₁G coordinate and h coordinate, g₂And h₂Respectively summit p₂G coordinate and h coordinate, g₃And h₃Respectively summit p₃G coordinate and h coordinate, g_rAnd h_rThe respectively g coordinate of output voltage reference vector and h coordinate.

2. method according to claim 1, is characterized in that the described vector sequence of operation determining residing subregion specifically, from subregion corresponding with vector sequence of operation storage organization is searched residing subregion press the vector identification that vector sequence of operation sorts; Further according to vector identification, vector identification storage organization is searched the vector that vector identification is corresponding.

3. method according to claim 2, is characterized in that described subregion and vector sequence of operation storage organization are array, chained list, class or tables of data.

4. method according to claim 2, is characterized in that described vector identification storage organization is array, chained list, class or tables of data.

5. the method according to any one claim in claim 2-4, is characterized in that the process of setting up of described vector identification storage organization is:

Step A1: set up the first storage organization, stores in the first storage organization by the sector mark belonging to each subregion, partition identification and each summit of the subregion coordinate in g-h coordinate system;

Step A2: by g-h coordinate system to two units of right translation, obtain g '-h ' coordinate system;

Step A3: set up the second storage organization, stores in the second storage organization by the sector mark belonging to each subregion, partition identification and each summit of the subregion coordinate in g '-h ' coordinate system;

Step A4: set up the 3rd storage organization, stores in the 3rd storage organization by the vector state on the summit of each subregion g ' coordinate in g '-h ' coordinate system, h ' coordinate and summit;

Step A5: set up vector identification storage organization, the vector on each each summit of subregion is extracted from the 3rd storage organization, and be that the vector of each extraction sets vector identification, then vector identification corresponding for vector described in the vector on each each summit of subregion is stored in vector identification storage organization.

6. method according to claim 5, is characterized in that described first storage organization, the second storage organization and the 3rd storage organization are array, chained list, class or tables of data.

7. method according to claim 6, the process of setting up of described subregion and vector sequence of operation storage organization that it is characterized in that is:

Step B1: determine the vector sequence of operation of each subregion, and according to described vector sequence of operation, vector identification is sorted;

Step B2: set up subregion and vector sequence of operation storage organization, stores the sector mark belonging to each subregion, partition identification and the vector identification sorted by vector sequence of operation in subregion and vector sequence of operation storage organization.