CN105262400A - Control method for improving compensation control accuracy of motor dead zone - Google Patents

Abstract

The invention relates to the air-conditioning control technique, and discloses a control method for improving the compensation control accuracy of a motor dead zone, which solves the acquisition problem of an arc tangent angle while improving the acquiring accuracy of the arc tangent angle so as to improve the compensation accuracy of the dead zone. The method comprises the steps of establishing a high-accuracy arc tangent data look-up table for a single-chip microcomputer to acquire the compensation data of the dead zone and compensate the dead zone, wherein the high-accuracy arc tangent data look-up table is subdivided into 2n parts based on 0-45 degree tangent function values and tangent values in each part are set apart; determining an angle value corresponding to each tangent function value and amplifying the angle value; rounding the amplified angle value and storing the rounded value in the ROM of the single-chip microcomputer for the single-chip microcomputer to inquire and use. According to the invention, generally, n is larger than or equal to 8, and m is larger than or equal to 16. The above method is used for controlling the permanent-magnet brushless direct current motor of an air conditioner.

Description

A kind of control method improving motor dead area compensation control precision

Technical field

The present invention relates to air conditioner controlling technology, particularly a kind of control method improving motor dead area compensation control precision.

Background technology

Permanent-magnet brushless DC electric machine controls generally to adopt vector control, when the direction of compute vectors, needs to calculate acquisition current vector to the angle of static coordinate axle α (or fixed coordinates axle u axle), namely needs arctangent computation to obtain.The control method of the dead area compensation that patent " a kind of dead time compensation control method and system (201310534766.3) " is introduced, also needs to obtain by I _d ^*, I _q ^*angle between the vector determined and U reference axis; determine symbol and the size of vector compensation; but do not provide concrete arc tangent and obtain control method; the dead time compensation control method based on arc tangent cannot be realized; test also finds, at three-phase current zero-acrross ing moment, due to control precision problem; easily cause the positive and negative of dead area compensation to occur mistake at random, cause overcurrent to shut down.

Summary of the invention

Technical problem to be solved by this invention is: propose a kind of control method improving motor dead area compensation control precision, while solving arc tangent angle acquisition problem, improves angle and obtains precision, and then improve dead area compensation precision.

The present invention solves the problems of the technologies described above adopted technical scheme,

Improving a control method for motor dead area compensation control precision, comprising: to table look-up data by setting up high-precision arc tangent, calculate for single-chip microcomputer and obtain dead area compensation numerical value, dead band is compensated;

Described high-precision arc tangent is tabled look-up data, and be by by 0 ~ 45 degree of tan value, being subdivided into is 2 ⁿpart, the tangent value of every part is separated by obtain the angle value that each tan value is corresponding, then angle value is amplified round, is stored in single-chip microcomputer ROM, inquires about for single-chip microcomputer; Generally get n>=8, m>=16.

Further, by obtaining the angle between current phasor and U axle, obtaining dead area compensation value, dead band is compensated.

Further, the method specifically comprises the following steps:

A. 0 ~ 45 degree of tan value is divided into 2 ⁿ, every part be i.e. interval set up data from 0, calculate the arc-tangent value set up, obtain the angle of its correspondence, adopt 360 ° of correspondences 2 ^mthe relation of numerical value, carry out amplification process to data, result of calculation adopts the method rounded up to obtain corresponding integer, and off-line sets up table [2 ⁿ+ 1] show, be stored in single-chip microcomputer ROM, for single-chip microcomputer inquiry, n is larger, and precision is higher, generally gets n>=8; M is larger, and precision is also higher, generally gets m>=16;

B. obtain by I _d*/I _q* determined current phasor with the included angle of d axle:

When be in d/q coordinate system first quartile, if I _d*>=I _q* time, if I _q* >I _d* time, especially, I is worked as _q*=I _dwhen *=0, φ=0;

When be in the second quadrant, I _q*>=| I _d* | time, if I _q* <|I _d* | time, $\mathrm{\&phi;}=2^{m-1}-table\&lsqb;arctan\left(\frac{2^n\&CenterDot;I_q*}{\left|I_d*\right|}\right)\&rsqb;;$

When be in third quadrant, if | I _q* |>=| I _d* | time, if | _q* | <|I _d* | time, $\mathrm{\&phi;}=2^{m-1}+table\&lsqb;arctan\left(\frac{2^n\&CenterDot;\left|I_q*\right|}{\left|I_d*\right|}\right)\&rsqb;;$

When be in fourth quadrant, if | I _q* |>=I _d* time, if | I _q* | <I _d* time, $\mathrm{\&phi;}=2^m-table\&lsqb;arctan\left(\frac{2^n\&CenterDot;I_d*}{\left|I_q*\right|}\right)\&rsqb;;$

C. the angle theta between d axle and fixed coordinates axle U is obtained;

D. according to current phasor with the position residing for the angle (θ+φ) of U axle, determine the size of dead area compensation:

When $0\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{2^{m-2}}{3}$ Time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2};$

When $\frac{2^{m-2}}{3}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2};$

When $2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{5}{3}\&CenterDot;2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2};$

When $\frac{5}{3}\&CenterDot;2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{7}{3}\&CenterDot;2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=\frac{T_d}{2};$

When $\frac{7}{3}\&CenterDot;2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<3\&CenterDot;2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=\frac{T_d}{2};$

When $3\&CenterDot;2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{11}{12}\&CenterDot;2^m$ Time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=\frac{T_d}{2};$

When time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2}.$

The invention has the beneficial effects as follows: by above-mentioned a kind of control method improving motor dead area compensation control precision, not only can obtain arctangent computation data, and current phasor more accurately can be obtained angle (θ+φ) data with U axle, obtain more accurate dead area compensation, improve current waveform, improve system overall efficiency, reduce system cloud gray model noise.

Accompanying drawing illustrates:

Fig. 1 is vector 45 degree of schematic diagrames are less than with horizontal axis angle;

Fig. 2 is vector 45 degree of schematic diagrames are greater than with horizontal axis angle;

Fig. 3 is current phasor schematic configuration diagram under three-coordinate system;

Fig. 4 is U/V/W three-phase current waveform schematic diagram.

Embodiment

The present invention is intended to propose a kind of control method improving motor dead area compensation control precision, while solving arc tangent angle acquisition problem, improves angle and obtains precision, and then improve dead area compensation precision.

Because motor angle controls to be generally 0 ~ 360 °, by analyzing, can be subdivided into and ask 0 ~ 45 °, then obtain the angle of 0 ~ 360 ° by algorithm, ask vector and during angle theta between x reference axis, due to as shown in Figure 1, and tan (45 °)=1, can adopt and carry out control acquisition with the following method:

By 0 ~ 45 degree of tan value, being subdivided into is 2 ⁿpart, the tangent value of every part is separated by 2 are divided into by 1 ⁿ, every part be i.e. interval set up data from 0, calculate the arc-tangent value set up, obtain the angle (unit is degree, or radian) of its correspondence, the angle due to its correspondence is decimal, in order to improve its precision, when calculated angular unit is for spending, can adopt 360 ° of correspondences 2 ^mthe relation of numerical value, carries out amplification process to data; When calculated angular unit is radian, 2 π correspondences 2 can be adopted ^mthe relation of numerical value, carry out amplification process to data, result of calculation adopts the method rounded up to obtain corresponding integer, sets up a data form table [2 ⁿ+ 1], to table look-up process for single-chip microcomputer.Work as n=8, during m=16, set up 2 ⁿ+ 1 data form table [257]=and 0,41,81,122,163 ..., 8110,8131,8151,8172,8192}.

When be in first quartile, and during Rx>=Ry>=0, as shown in Figure 1, due to Rx>=Ry>=0, so during owing to setting up form, be divided into 2 by 1 ⁿ, so will as lookup table index, inquiry table [2 ⁿ+ 1] form, obtains when angle process below, notice that angle needs again according to 360 ° of correspondences 2 ^mor 2 π correspondences 2 ^mnumerical relation, reprocessing, also can by table look-up data divided by or divided by reduction is tabled look-up angular values, determine according to follow-up angle processing method, and " " wherein represents multiplying, lower together.

When be in first quartile, and during Ry>Rx>0, θ=90-γ, namely or $\mathrm{\&theta;}=\frac{90}{360}\&CenterDot;2^m-table\&lsqb;arctan\left(\frac{2^n\&CenterDot;Rx}{Ry}\right)\&rsqb;=2^{m-2}-table\&lsqb;arctan\left(\frac{2^n\&CenterDot;Rx}{Ry}\right)\&rsqb;,$ As shown in Figure 2.

Thus, Angle ambiguity algorithm is obtained as follows:

When be in first quartile, if during Rx>=Ry, if during Ry>Rx, especially, as Rx=Ry=0, θ=0;

When be in the second quadrant, Ry>=| during Rx|, if during Ry<|Rx|, $\mathrm{\&theta;}=2^{m-1}-table\&lsqb;arctan\left(\frac{2^n\&CenterDot;Ry}{\left|Rx\right|}\right)\&rsqb;;$

When be in third quadrant, if | Ry|>=| during Rx|, if | during Ry|<|Rx|, $\mathrm{\&theta;}=2^{m-1}+table\&lsqb;arctan\left(\frac{2^n\&CenterDot;\left|Ry\right|}{\left|Rx\right|}\right)\&rsqb;;$

When be in fourth quadrant, if | during Ry|>=Rx, if | during Ry|<Rx, $\mathrm{\&theta;}=2^m-table\&lsqb;arctan\left(\frac{2^n\&CenterDot;\left|Ry\right|}{Rx}\right)\&rsqb;;$

The present invention adopts coordinate system as shown in Figure 3, in figure, U, V, W three-phase difference is 120 °, α/β is static rectangular coordinate system, d/q is the rectangular coordinate system along with rotor rotates, α and U phase direction is consistent, wherein θ is the angle between d axle and α axle, is also the position angle of rotor for motor actual current command vector, by current order vector I _d ^*, I _q ^*uniquely determine, I _d ^*and I _q ^*be respectively the order of d shaft current and the order of q shaft current.φ is electric current and the angle between d axle.Will project in U, V, W reference axis, obtain I respectively _u ^*, I _v ^*, I _w ^*.Dead area compensation control unit is according to I _u ^*, I _v ^*, I _w ^*positive and negative, determine dead area compensation △ T _u, △ T _v, △ T _wpositive and negative and size, wherein △ T _ufor producing PWM ripple duty cycle register T _uneed the size compensated, △ T _vfor producing PWM ripple duty cycle register T _vneed the size compensated, △ T _wfor producing PWM ripple duty cycle register T _wneed the size compensated.Work as I _u ^*during >0, i _u ^*during <0 work as I _v ^*during >0, i _v ^*during <0 work as I _w ^*during >0, time, I _w ^*during <0 and work as I _u ^*=0, or , be from positive to negative or from negative to positive depending on the sense of current, I _v ^*=0, I _w ^*when=0, principle is identical, can compensate just also can compensate negative.

Therefore, the present invention is by current order vector i.e. I _d ^*, I _q ^*determine I _u ^*, I _v ^*, I _w ^*the method of sign is:

Due to ${I_u}^{*}=\left|\stackrel{\&RightArrow;}{I^{*}}\right|cos(\mathrm{\&theta;}+\mathrm{\&phi;}),{I_v}^{*}=\left|\stackrel{\&RightArrow;}{I^{*}}\right|cos(\mathrm{\&theta;}+\mathrm{\&phi;}-120),{I_w}^{*}=\left|\stackrel{\&RightArrow;}{I^{*}}\right|cos(\mathrm{\&theta;}+\mathrm{\&phi;}+120),$ Wherein represent electric current amplitude, the scope only residing for angle (θ+φ) I can be determined _u ^*, I _v ^*, I _w ^*positive and negative; As 0≤(θ+φ) <30 °, I _u ^*>0, I _v ^*<0, I _w ^*<0.Because the angle theta between d axle and α axle is obtained by other program control, size and the positive and negative problem of dead area compensation are converted into the size asking φ;

The method of angle φ is asked to be:

When be in d/q coordinate system first quartile, if I _d*>=I _q* time, if I _q* >I _d* time, especially, I is worked as _q*=I _dwhen *=0, φ=0;

When be in the second quadrant, I _q*>=| I _d* | time, if I _q* <|I _d* | time, $\mathrm{\&phi;}=2^{m-1}-table\&lsqb;arctan\left(\frac{2^n\&CenterDot;I_q*}{\left|I_d*\right|}\right)\&rsqb;;$

When be in third quadrant, if | I _q* |>=| I _d* | time, if | I _q* | <|I _d* | time, $\mathrm{\&phi;}=2^{m-1}+table\&lsqb;arctan\left(\frac{2^n\&CenterDot;\left|I_q*\right|}{\left|I_d*\right|}\right)\&rsqb;;$

When be in fourth quadrant, if | I _q* |>=I _d* time, if | I _q* | <I _d* time, $\mathrm{\&phi;}=2^{m-1}-table\&lsqb;arctan\left(\frac{2^n\&CenterDot;I_d*}{\left|I_q*\right|}\right)\&rsqb;;$

As seen from Figure 4, when θ, φ unit is degree, then compensate according to following relation:

As 0≤(θ+φ) <30 °, I _u ^*>0, I _v ^*<0, I _w ^*<0,

As 30 °≤(θ+φ) <90 °, I _u ^*>0, I _v ^*>=0, I _w ^*<0,

As 90 °≤(θ+φ) <150 °, I _u ^*≤ 0, I _v ^*>0, I _w ^*<0,

As 150 °≤(θ+φ) <210 °, I _u ^*<0, I _v ^*>0, I _w ^*>=0,

As 210 °≤(θ+φ) <270 °, I _u ^*<0, I _v ^*≤ 0, I _w ^*>0,

As 270 °≤(θ+φ) <330 °, I _u ^*>=0, I _v ^*<0, I _w ^*>0,

As 330 °≤(θ+φ) <360 °, I _u ^*>0, I _v ^*<0, I _w ^*≤ 0,

According to 360 ° of correspondences 2 ^m, known 30 ° of correspondences 90 ° of correspondences 2 ^m-2, 150 ° of correspondences 210 ° of correspondences 270 ° of correspondences 32 ^m-2, 330 ° of correspondences when θ, φ are according to according to 360 ° of correspondences 2 ^mduring relation determined value, then compensate according to following relation:

When $0\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{2^{m-2}}{3}$ Time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2};$

When $\frac{2^{m-2}}{3}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2};$

When $2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{5}{3}\&CenterDot;2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2};$

When $\frac{5}{3}\&CenterDot;2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{7}{3}\&CenterDot;2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=\frac{T_d}{2};$

When $\frac{7}{3}\&CenterDot;2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<3\&CenterDot;2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=\frac{T_d}{2};$

When $3\&CenterDot;2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{11}{12}\&CenterDot;2^m$ Time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=\frac{T_d}{2};$

When time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2};$

Wherein θ, φ are according to according to 360 ° of correspondences 2 ^mrelation determined value; Visible, when θ, φ are according to according to 360 ° of correspondences 2 ^mduring relation determined value, θ, φ expand doubly, as m=16, then θ, φ expand 182 times; When θ, φ are according to according to 2 π correspondences 2 ^mduring relation determined value, θ, φ expand doubly, as m=16, then θ, φ expand 10430 times, compensate and accurately significantly improve.

On concrete enforcement, the control method of the raising motor dead area compensation control precision in the present invention, comprises the steps:

A. 2 are divided into by 1 ⁿ, every part be i.e. interval set up data from 0, calculate the arc-tangent value set up, obtain the angle of its correspondence, adopt 360 ° of correspondences 2 ^mthe relation of numerical value, carries out amplification process to data, and calculate result and adopt the method rounded up to obtain corresponding integer, off-line sets up table [2 ⁿ+ 1] show, be stored in MCUROM, for single-chip microcomputer inquiry, n is larger, and precision is higher, generally gets n>=8; M is larger, and precision is also higher, generally gets m>=16;

B. obtain by I _d*/I _q* determined current phasor with the included angle of d axle:

When be in d/q coordinate system first quartile, if I _d*>=I _q* time, if I _q* >I _d* time, especially, I is worked as _q*=I _dwhen *=0, φ=0;

When be in the second quadrant, I _q*>=| I _d* | time, if I _q* <|I _d* | time, $\mathrm{\&phi;}=2^{m-1}-table\&lsqb;arctan\left(\frac{2^n\&CenterDot;I_q*}{\left|I_d*\right|}\right)\&rsqb;;$

When be in third quadrant, if | I _q* |>=| I _d* | time, if | I _q* | <|I _d* | time, $\mathrm{\&phi;}=2^{m-1}+table\&lsqb;arctan\left(\frac{2^n\&CenterDot;\left|I_q*\right|}{\left|I_d*\right|}\right)\&rsqb;;$

When be in fourth quadrant, if | I _q* |>=I _d* time, if | I _q* | <I _d* time, $\mathrm{\&phi;}=2^m-table\&lsqb;arctan\left(\frac{2^n\&CenterDot;I_d*}{\left|I_q*\right|}\right)\&rsqb;;$

C. the angle theta between d axle and fixed coordinates axle U is obtained;

D. according to current phasor with the position residing for the angle (θ+φ) of U axle, determine the size of dead area compensation and positive and negative:

When $0\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{2^{m-2}}{3}$ Time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2};$

When $\frac{2^{m-2}}{3}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2};$

When $2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{5}{3}\&CenterDot;2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2};$

When $\frac{5}{3}\&CenterDot;2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{7}{3}\&CenterDot;2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=\frac{T_d}{2};$

When $\frac{7}{3}\&CenterDot;2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<3\&CenterDot;2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=\frac{T_d}{2};$

When $3\&CenterDot;2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{11}{12}\&CenterDot;2^m$ Time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=\frac{T_d}{2};$

When time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2}.$

Claims (3)

1. improving a control method for motor dead area compensation control precision, it is characterized in that, comprising: to table look-up data by setting up high-precision arc tangent, calculate for single-chip microcomputer and obtain dead area compensation numerical value, dead band is compensated;

Described high-precision arc tangent is tabled look-up data, and be by by 0 ~ 45 degree of tan value, being subdivided into is 2 ⁿpart, the tangent value of every part is separated by obtain the angle value that each tan value is corresponding, then angle value is amplified round, is stored in single-chip microcomputer ROM, inquires about for single-chip microcomputer; Generally get n>=8, m>=16.

2. a kind of control method improving motor dead area compensation control precision as claimed in claim 1, is characterized in that, by obtaining the angle between current phasor and U axle, obtaining dead area compensation value, compensating dead band.

3. a kind of control method improving motor dead area compensation control precision as claimed in claim 2, it is characterized in that, the method specifically comprises the following steps:

A. 0 ~ 45 degree of tan value is divided into 2 ⁿ, every part be i.e. interval set up data from 0, calculate the arc-tangent value set up, obtain the angle of its correspondence, adopt 360 ° of correspondences 2 ^mthe relation of numerical value, carry out amplification process to data, result of calculation adopts the method rounded up to obtain corresponding integer, and off-line sets up table [2 ⁿ+ 1] show, be stored in single-chip microcomputer ROM, for single-chip microcomputer inquiry, described n>=8, m>=16;

B. obtain by I _d*/I _q* determined current phasor with the included angle of d axle:

When be in d/q coordinate system first quartile, if I _d*>=I _q* time, if I _q* >I _d* time, especially, I is worked as _q*=I _dwhen *=0, φ=0;

When be in the second quadrant, I _q*>=| I _d* | time, if I _q* <|I _d* | time, $\mathrm{\&phi;}=2^{m-1}-table\&lsqb;arctan\left(\frac{2^n\&CenterDot;I_q*}{|I_d*|}\right)\&rsqb;;$

When be in third quadrant, if | I _q* |>=| I _d* | time, $\mathrm{\&phi;}=3\&CenterDot;2^{m-2}-table\&lsqb;arctan\left(\frac{2^n\&CenterDot;|I_d*|}{|I_q*|}\right)\&rsqb;,$ If | I _q* | <|I _d* | time, $\mathrm{\&phi;}=2^{m-1}+table\&lsqb;arctan\left(\frac{2^n\&CenterDot;|I_q*|}{|I_d*|}\right)\&rsqb;;$

When be in fourth quadrant, if | I _q* |>=I _d* time, if | I _q* | <I _d* time, $\mathrm{\&phi;}=2^m-table\&lsqb;arctan\left(\frac{2^n\&CenterDot;I_d*}{|I_q*|}\right)\&rsqb;;$

C. the angle theta between d axle and fixed coordinates axle U is obtained;

D. according to current phasor with the position residing for the angle (θ+φ) of U axle, determine the size of dead area compensation:

When $0\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{2^{m-2}}{3}$ Time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2};$

When $\frac{2^{m-2}}{3}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2};$

When $2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{5}{3}\&CenterDot;2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2};$

When $\frac{5}{3}\&CenterDot;2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{7}{3}\&CenterDot;2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=\frac{T_d}{2};$

When $\frac{7}{3}\&CenterDot;2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<3\&CenterDot;2^{m-2}$ Time, ${\mathrm{\&Delta;T}}_u=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=\frac{T_d}{2};$

When $3\&CenterDot;2^{m-2}\&le;(\mathrm{\&theta;}+\mathrm{\&phi;})<\frac{11}{12}\&CenterDot;2^m$ Time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=\frac{T_d}{2};$

When time, ${\mathrm{\&Delta;T}}_u=\frac{T_d}{2},{\mathrm{\&Delta;T}}_v=-\frac{T_d}{2},{\mathrm{\&Delta;T}}_w=-\frac{T_d}{2}.$