JP2005210764A - Drive controller of brushless motor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a drive controller of a brushless motor which can suppress certainly a current change, a torque change and generation of a noise in a brushless motor, and which can build a motor-driven power steering apparatus having good steering feeling in a low noise when applied as a torque assistant device of the motor-driven power steering apparatus. <P>SOLUTION: The drive controller of the brushless motor includes a plurality of excitation phases driven by a rectangular wave-like excitation current. The drive controller of the brushless motor further includes a current instruction value calculating means for calculating a current instruction value to the brushless motor, and a commutation pattern generating part for generating a commutation pattern of each phase given by a predetermined numerical value beforehand set at the commutation time. Further, each phase current instruction value calculating means for calculating a current instruction value to each phase of the brushless motor based on the commutation pattern and the current instruction value is provided. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a drive control device for a brushless motor having a plurality of excitation phases, and more particularly to a drive control device for a brushless motor suitable for a drive source of an electric power steering device.

  A brushless motor used as a drive source for a power steering device of an automobile is a motor having three or more excitation phases, and is driven by a rectangular wave-like excitation current.

  For example, in the case of a five-phase brushless motor, the motor drive circuit has five phases (hereinafter referred to as a phase to e phase) arranged so as to surround the outer circumferential surface of the rotor of the motor with an electrical angle of 72 degrees apart. The excitation coils a to e are controlled by a control circuit such as a microcomputer, and the four phases are excited simultaneously, and the coils are sequentially switched one by one and excited by a rectangular wave current. It is driven to rotate. In this four-phase excitation method, the motor current flows in four of the five phases, but in order to flow the current in a balanced manner, the resistances of the respective excitation coils are all made equal. Yes.

  Such a motor drive circuit is usually composed of ten field effect transistors (FETs). These 10 transistors are formed by connecting two corresponding transistors in series to form five series transistor circuits, each being connected between the positive and negative terminals of the power source, and the two transistors of each series transistor circuit. By connecting the connecting portions to the outer ends of the five exciting coils a to e that are star-connected in a Y shape, they are connected to the motor coil circuit.

  The direction and length of the excitation current (rectangular wave) supplied from the motor drive circuit to each excitation coil are as shown in FIG. 8, for example, with respect to the value of the rotation angle (electrical angle) of the rotor. That is, the rotor is continuously rotated by switching the excitation coil one phase at a time every 36 degrees in electrical angle and exciting one phase coil for 144 degrees in electrical angle. In FIG. 8, when the electrical angle is θ, 0 ° ≦ θ <36 °, 36 ° ≦ θ <72 °, 72 ° ≦ θ <108 °, 108 ° ≦ θ <144 °, 144 ° ≦ θ < 180 °, 180 ° ≦ θ <216 °, 216 ° ≦ θ <252 °, 252 ° ≦ θ <288 °, 288 ° ≦ θ <324 °, 324 ° ≦ θ <360 ° (1),・ ・ ・ ・, (10)

  In this example, the a-phase current flows in the positive direction in the sections (1) and (2), 0 in the section (3), in the negative direction in the sections (4) to (7), and in the section (8). 0, it flows in the positive direction again in the section (1) through the sections (9) to (10). The b-phase current flows in the positive direction in sections (1) to (4), 0 in section (5), negative in sections (6) to (9), 0 in section (10), and again It flows in the positive direction in section (1). The c-phase current flows in the negative direction in section (1), 0 in section (2), flows in the positive direction in sections (3)-(6), 0 in section (7), and sections (8)-( After 10), it flows again in the negative direction in section (1). The d-phase current flows in the negative direction in the sections (1) to (3), 0 in the section (4), in the positive direction in the sections (5) to (8), 0 in the section (9), and the section It flows again in the negative direction from (10). The e-phase current is 0 in the section (1), flows in the negative direction in the sections (2) to (5), 0 in the section (6), flows in the positive direction in the sections (7) to (10), and again in the section It becomes 0 in (1). Therefore, two of the five exciting coils are switched in the opposite directions at each boundary of the sections (1) to (10) (when the electrical angle is switched every 36 degrees).

Such switching of the excitation current is theoretically expressed by a rising or falling edge of a rectangular wave as shown in FIG. 8, but actually the rising or falling waveform is shown on the horizontal axis (time or rotation angle). However, it takes a certain time Δt (three times the time constant of the motor circuit) until the exciting current rises in the positive direction or falls in the negative direction. For example, at the boundary between sections (8) and (9) in FIG. 8 (electrical angle 288 degrees), the a-phase current rises from 0 to a positive constant value, while the d-phase current rises from a positive constant value to 0. The b-phase and c-phase currents are both “negative” constant values, and the e-phase current is “positive” constant values. When the change in the waveform at this boundary portion is enlarged, as shown in FIG. become. When the currents of the five phases are represented by i _a , i _b , i _c , i _d and i _e , there is a relationship of the following formula (1) between these currents.

i _a + i _d + i _e = − (i _b + i _c ) = I (1)

For this reason, for example, when the a-phase and d-phase currents change, the b-, c-, and e-phase currents also change. In this case, since the current change rates of the a phase and the d phase are different, the total value of the currents of the two phases does not become a steady value, and the currents of the b phase and the c phase fluctuate as shown in FIG. The current also changes during the time Δt. These current fluctuations cause excessive torque fluctuations.

  The reason why the current change rates of the rising and falling currents of the two phases are different is as follows.

  First, let Vb be the power supply voltage supplied to the motor drive circuit, and let Vn be the voltage at the center connection point of the excitation coils a to e that are star-connected. Next, in FIG. 9, the section of time Δt1 is (a), and the section of time Δt2 (= Δt−Δt1) is (b).

In the section (a), the d-phase (OFF-phase) current i _{d that} is switched from positive to zero has a rate of change corresponding to −Vn, the back electromotive force E _{d of the} coil, and the time constant of the motor circuit. Decreases from half (I / 2) of the current I applied to the motor to zero (0). At this time, if the voltage applied to the OFF-phase equivalent circuit is V _OFF , V _OFF = −Vn−Ed <0, and Vn is approximately Vb / 2. On the other hand, the current i _a of a phase which is positively switched from 0 (ON phase), increase the voltage Vb, -Vn, the change rate according to the time constant of the back electromotive force Ea and the motor circuit of the coil from zero (0) However, if the voltage applied to the ON-phase equivalent circuit is V _{ON at} this time, V _ON = Vb · Duty1 (PWM duty) −Vn−Ea.

Explained by the equation, the current i _d is expressed by the following equation (2) by an equivalent circuit of the OFF phase.

i _d (t) = (I / 2) e ^{−t / T} + (V _OFF / R) (1−e ^{−t / T} ) (2)
When ∴t = 0, i _d = I / 2
Where T is the electrical time constant of the equivalent circuit, and R is the resistance of the equivalent circuit.

On the other hand, the current i _a is expressed by the following equation (3) by an ON-phase equivalent circuit.

i _a (t) = (V _ON / R) (1−e ^{−t / T} ) (3)
When ∴t = 0, i _a = 0, t → ∞, i _a = V _ON / R = I / 2

Thus, OFF-phase, the current i _d of ON phase, the rate of change of i _a, respectively as follows.

di _d (t) / dt =-(1 / T) (I / 2) e- ^{t / T} + (1 / T) ( _VOFF / R) e- ^{t / T}
=-(I / 2-V _OFF / R) (1 / T) e ^{-t / T}
=-(I / 2 + Vn / R + Ed / R) (1 / T) e- ^{t / T} (4)
di _a (t) / dt = (1 / T) (V _ON / R) e ^{-t / T}
= (I / 2) (1 / T) e ^{-t / T} (5)

In the above equations (4) and (5), since (I / 2 + Vn / R + Ed / R)> I / 2, the OFF phase current change rate is larger than the ON phase current change rate. In particular, when the resistance R of the equivalent circuit is small, the power supply voltage Vb (≈2Vn) is large, or the counter electromotive voltage Ed is large at high speed rotation, the OFF phase current change rate is higher than the ON phase current change rate. It gets quite big. Therefore, the time (Δt) during which the ON-phase current i _a rises from 0 to I / 2 is longer than the time (Δt1) during which the OFF-phase current i _d falls from I / 2 to 0. Thereafter, in the section (b), the ON-phase current i _a finally reaches the steady value (I / 2), but it takes time Δt2 (2 to 3 times the time constant of the motor circuit). Therefore, the current change rate differs between the rising and falling currents of the two phases to be switched.
JP 2001-268879 A

  As described above, in the excitation current control by the conventional motor drive circuit, since the rate of change of the rising and falling currents of the two phases to be switched (for example, the a phase and the d phase in FIG. 8) are different, the phases that cannot be switched ( For example, the currents of b-phase, c-phase, and e-phase in FIG.

  As a solution to such a problem, there is one disclosed in Patent Document 1 (Japanese Patent Laid-Open No. 2001-268879). In Patent Document 1, in a drive control device for a brushless motor that has a current change rate control unit that controls a current change rate during commutation and has a plurality of excitation phases, the current change rate control unit includes a phase change in a commutation operation. The commutation transient time, which is the time during which the current is in a transient state, is terminated within half the commutation interval time, which is the time required from the start time of one commutation to the start time of the next commutation. I am doing so.

  However, as described above, the rising phase current and the falling phase current are different, and the phase current that is not commutated changes. Therefore, the sum of the phase currents becomes 0, and the phase current becomes a transient state in the commutation operation. If the motor is driven and controlled so that the transition time of the commutation is completed within half of the commutation interval time required from the start time of one commutation to the start time of the next commutation, vibration or There is a problem that noise is generated.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to reliably suppress the occurrence of current fluctuation, torque fluctuation and noise in a brushless motor, and a torque assist device for an electric power steering apparatus. The present invention is to provide a drive control device for a brushless motor capable of constructing an electric power steering device with low noise and good steering feeling.

  The present invention relates to a drive control device for a brushless motor having a plurality of excitation phases driven by a rectangular wave excitation current, and the object of the present invention is to provide a current command value calculation means for calculating a current command value for the brushless motor. And a commutation pattern generator for generating a commutation pattern of each phase given by a predetermined numerical value at the time of commutation, and to each phase of the brushless motor based on the commutation pattern and a current command value This is achieved by providing each phase current command value calculating means for calculating the current command value.

  The above object of the present invention is a set of numerical values in which the sum of the two phases of the commutation pattern is 1 or −1, and the current to each phase is multiplied by the commutation pattern and the current command value. This is achieved more effectively by calculating a command value or by making the commutation pattern into a sine wave waveform.

  According to the drive control device for a brushless motor of the present invention, the sum of the two phases of the commutation pattern at the time of commutation is a set of numerical values that becomes 1 or −1. Since the current command value to the phase is calculated and the commutation pattern is given as a sinusoidal waveform, the current command value with low vibration and noise of the brushless motor can be given. Therefore, according to the present invention, it is possible to reliably suppress the occurrence of current fluctuation, torque fluctuation and noise in the brushless motor.

  Further, if the brushless motor drive control device according to the present invention is used as an assist motor drive control device for an electric power steering device, a high-grade electric power steering device with low noise and good steering feeling can be constructed. .

  The brushless motor drive control device according to the present invention provides a current command value with a sine wave (or pseudo sine wave) commutation pattern generated during commutation in order to suppress generation of vibration and noise during commutation. . At this time, the sum of the two phases of the commutation pattern is a set of values such that 1 (integer is 1024) or -1 (integer is -1024), and each of the brushless motors is multiplied by the commutation pattern and the current command value. The current command value for the phase is calculated. As a result, even when the brushless motor is subjected to rectangular wave control, low vibration and low noise drive control can be realized.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  First, a motor drive control device 20 according to the present invention will be described with reference to FIG. 1 for a five-phase brushless motor. The motor drive control device 20 includes a control circuit 21, an FET gate drive circuit 22, a motor drive circuit 23, a current detection circuit 24, and a rotor position detection circuit 25. The control circuit 21 is constituted by a microcomputer, for example, and a constant voltage is supplied from the constant voltage source 26. The control circuit 21 receives the current command Iref from the external circuit 27, the motor current detection value I from the current detection circuit 24, and the rotor position signal Sa-e (= Sa,..., Se) from the rotor position detection circuit 25. Are entered respectively. The control circuit 21 controls the drive current supplied from the motor drive circuit 23 to the coil circuit 12 of the brushless motor 1 based on these input signals.

  Here, when the five-phase brushless motor 1 is used as a drive source of the electric power steering apparatus, the external circuit 27 is a vehicle speed obtained from an output of a vehicle speed sensor that generates a pulse signal corresponding to the output shaft rotational speed of the transmission of the automobile. A motor corresponding to the detected value V and a detected value T including the direction of the torque obtained from the output of the torque sensor that detects the steering torque applied to the input shaft of the steering wheel with reference to a predetermined characteristic diagram It is configured to search for a current value and output it as a current command Iref.

  The motor drive circuit 23 is composed of a total of 10 transistors (field effect transistor FETs) Ta1 to Te1, Ta2 to Te2 with 5 on the power supply side (upper side) and 5 on the ground side (lower side). . The ten transistors Ta1 to Te1 and Ta2 to Te2 have corresponding transistors connected in series on the upper stage side and the lower stage side, and these series-connected transistor pairs (Ta1-Ta2, Tb1-Tb2, Tc1-Tc2, Td1- The upper stage terminals of Td2, Te1-Te2) are connected to the control circuit 21, the lower stage terminals are connected to the current detection circuit 24, and the connection portions of the transistor pairs are the outer ends of the exciting coils 6a to 6e. (The opposite side to the center side of the star connection). The gate voltages of the transistors Ta1 to Te2 are controlled by the control circuit 21 based on the detection signal Sa-e from the rotor position detection circuit 25.

  The direction and magnitude of the excitation current from the motor drive circuit 23 to the respective excitation coils 6a to 6e are basically as shown in FIG. 8 as in the prior art, and the on / off timings of the transistors Ta1 to Te2 are as follows. The gate signals (upper stage) Ga1 to Ge1 and the gate signals (lower stage) Ga2 to Ge2 shown in Table 1 below are shown. In Table 1, gate signals Ga1 to Ge2 for turning on / off the transistors Ta1 to Te2 are represented by “1” and “0”, respectively.

図８において、ロータが例えば区間(1)の状態にあるものとすると、これは表１の区間(1)に該当し、上段側のトランジスタＴa1，Ｔb1及び下段側のトランジスタＴc2，Ｔd2がオン状態、これら以外のトランジスタはオフ状態であるので、励磁コイル６ａ及び６ｂには外端側から電流が流れ、励磁コイル６ｃ及び６ｄには結線側から電流が流れる。これにより、ロータのＮ極又はＳ極とその周囲に発生したＮ極又はＳ極との間に磁気吸引力及び反発力が生じ、ロータが回転する。そして、ロータが、図８の区間(2)の状態に移行すると、これは表１の区間(2)に該当し、上段側のトランジスタＴa1，Ｔb1及び下段側のトランジスタＴd2，Ｔe2がオン状態、これ以外のトランジスタがオフ状態であるので、励磁コイル６ａ及び６ｂには外端側から電流が流れ、励磁コイル６ｄ及び６ｅには結線側から電流が流れる。これにより、ロータが更に回転する。

In FIG. 8, if the rotor is in the state of section (1), for example, this corresponds to section (1) of Table 1, and the upper stage transistors Ta1 and Tb1 and the lower stage transistors Tc2 and Td2 are on. Since the other transistors are in the OFF state, current flows from the outer end side to the exciting coils 6a and 6b, and current flows from the connection side to the exciting coils 6c and 6d. Thereby, a magnetic attraction force and a repulsive force are generated between the N pole or S pole of the rotor and the N pole or S pole generated around the rotor, and the rotor rotates. When the rotor shifts to the state of section (2) in FIG. 8, this corresponds to section (2) of Table 1, and the upper stage transistors Ta1 and Tb1 and the lower stage transistors Td2 and Te2 are on. Since the other transistors are in the OFF state, current flows from the outer end side to the exciting coils 6a and 6b, and current flows from the connection side to the exciting coils 6d and 6e. As a result, the rotor further rotates.

  When the above operation is repeated, each transistor is driven at the timing shown in Table 1. As shown in FIG. 8, the excitation coil is sequentially switched by one phase every 36 degrees in electrical angle as shown in FIG. Energize for degrees. Thereby, the N pole or S pole generated in the stator sequentially moves, and the rotor continuously rotates. The FET gate drive circuit 22 supplies a predetermined voltage to the gate terminal of the designated transistor based on the gate drive signal G1-10 output from the control circuit 21.

  On the other hand, an example of a brushless motor (5-phase brushless motor 1) driven and controlled by the motor drive control device 20 is as shown in FIG. 2, and this 5-phase brushless motor 1 includes a cylindrical housing 2 and a housing 2 The rotating shaft 4 is disposed along the axis of the rotating shaft 4 and is rotatably supported by the bearings 3a and 3b. The motor driving permanent magnet 5 fixed to the rotating shaft 4 and the permanent magnet 5 are surrounded. The stator 6 is fixed to the inner peripheral surface of the housing 2 and wound with five-phase exciting coils 6a, 6b, 6c, 6d and 6e, and the rotor 7 is constituted by the rotating shaft 4 and the permanent magnet 5. ing.

  A ring-shaped permanent magnet 8 for phase detection is fixed in the vicinity of one end portion of the rotating shaft 4 of the rotor 7, and the permanent magnet 8 is alternately magnetized with S and N poles at equal intervals in the circumferential direction. Yes. The permanent magnet 5 of the rotor 7 is also magnetized with S poles and N poles alternately at equal intervals in the circumferential direction. On the end surface of the housing 2 on the side where the bearing 3b is disposed, a support substrate 10 made of a ring-shaped thin plate is disposed via a stay 9 so that the inner insulating portion faces the permanent magnet 8. ing. On the surface of the support substrate 10 on the permanent magnet 8 side, a phase detection element (hall sensor) 11 made of, for example, a hall element is fixed so as to face the permanent magnet 8. In practice, five phase detection elements 11 (11a to 11e) are provided at appropriate intervals in the circumferential direction corresponding to the drive timing of the excitation coils 6a to 6e. FIG. 2 shows one of them. Only shows.

  The phase detection elements 11a to 11e have an "H" sensor signal as a position detection signal when the magnetic poles of the permanent magnets 8 facing each other are N poles, and "L" sensor signals when they are S poles. Are output respectively. The rotational position of the rotor 7 can be detected by utilizing the fact that the outputs of the phase detection elements 11a to 11e change depending on the magnetic poles of the permanent magnet 8 facing the elements. Depending on the rotational position, the motor drive control device 20 energizes the five-phase excitation coils 6a to 6e simultaneously with four phases, while switching the energized excitation coils one by one in sequence, and thereby the rotor 7 is driven. It is designed to rotate.

  Further, the five-phase exciting coils 6a to 6e are disposed so as to surround the outer peripheral surface of the rotor 7 with an electrical angle of 72 degrees apart, and are star-connected in a Y shape as shown in FIG. A coil circuit 12 of the motor is configured. In the four-phase excitation method, the motor current flows in four phases, but since the current is inversely proportional to the coil resistance, the coil resistance of each of the excitation coils 6a to 6e is set to flow the current in a balanced manner. All are formed to be equal. The stator 6 has, for example, 30 slots at equal intervals on the inner peripheral surface of a stator core (not shown), and has the same number of convex portions between these slots. The exciting coils 6a to 6e are wound around. One end of each exciting coil 6 a to 6 e is connected together, and the other end is connected to the motor drive control device 20.

  The present invention generates commutation patterns 1 to 4 (see FIG. 8) according to a predetermined condition at the time of commutation by a commutation pattern generation unit provided in a control circuit 21 in a motor drive control device 20, and the current of each phase. The command values Irefas, Irefbs, and Irefcs are given to the brushless motor 1, and a configuration example is shown in FIG. In addition, although the brushless motor 1 of FIG. 1 has five phases, commutation will be described in three phases.

In the present invention, first, a current command value at the time of commutation based on a sine wave waveform is defined. At this time, whether the current command value of each phase is positive, negative, or 0 is determined by the position of the motor. For example, when there is no commutation, a positive / negative or zero current command value is calculated as follows.

1 × 2 ¹⁰ → Ireft2 × (1 × 2 ¹⁰ )
→ Iret2 × (1 × 2 ¹⁰ ) / 2 ¹⁰ → ・ Normal phase current command value

0 × 2 ¹⁰ → Ireft2 × (0 × 2 ¹⁰ )
→ Iret2 × (0 × 2 ¹⁰ ) / 2 ¹⁰ → ・ Zero phase current command value

−1 × 2 ¹⁰ → Ireft2 × (−1 × 2 ¹⁰ )
→ Iret2 × (-1 × 2 ¹⁰ ) / 2 ¹⁰ → ・ ・ Current command value of negative phase

That is, when it is not commutation, the current command value of the three phases is Ireft2, 0, −Ireft2.

For commutation, the calculation method is the same as for noncommutation. However, the value of multiplication is not 1, 0, -1, but is multiplied by the value of the sine wave waveform by the elapsed time. Here, an example of 12 points is shown. If the commutation is commutation from the zero phase to the positive phase, after 12 × sampling time has elapsed, the value of multiplication changes from 0 to 1, and the value therebetween is 0,0.017037,..., 0.98296. , 1 The current command value is calculated as follows.

Before commutation ia = [Ireft2 × (0 × 2 ¹⁰ )] / 2 ¹⁰
Step 1 ia = [Ireft2 × (0.01737 × 2 ¹⁰ )] / 2 ¹⁰
・
・
・
Step 11 ia = [Iref2 × (0.98296 × 2 ¹⁰ )] / 2 ¹⁰
Step 12 ia = [Iref2 × (1 × 2 ¹⁰ )] / 2 ¹⁰

In the commutation pattern generation unit, Ca, Cb, and Cc as parameters representing commutation are input to the commutation pattern generation circuits 211, 212, and 213, respectively, and Z- ¹ circuits 214, 215, and 215 in digital control, respectively. 216 (one sampling delay) is input to the commutation pattern generation circuits 211, 212, and 213. The current command value Ireft2 is input to the multiplication circuit, multiplied by the outputs of the commutation pattern generation circuits 211, 212, and 213, and the multiplication results are passed through the coefficient circuits ( ²⁻¹⁰ ) 231, 232, and 233, respectively. The values are output as Irefas, Irefbs, and Irefcs.

  The commutation pattern generation circuits 211, 212, and 213 generate four types of commutation patterns as shown in FIG. That is, the commutation pattern 1 has a sine wave waveform of “0 → 1”, the commutation pattern 2 has a sine wave waveform of “−1 → 0”, and the commutation pattern 3 has a sine wave of “1 → 0”. The commutation pattern 4 is a sine wave waveform of “0 → −1”. The commutation patterns 1 to 4 may be pseudo sine wave waveforms. Each of the commutation pattern generation circuits 211, 212, and 213 generates and outputs the commutation patterns 1 to 4 in accordance with predetermined input conditions.

  The commutation parameters Ca, Cb, and Cc are generated based on the phase detection element (Hall sensor), and are generated and output in the relationship shown in Table 2.

上述のような関係で生成出力される転流パラメータＣａ，Ｃｂ，Ｃｃが、それぞれ転流パターン発生回路２１１，２１２，２１３に入力される。転流パターン発生回路２１１，２１２，２１３は転流パラメータＣａ，Ｃｂ，Ｃｃの入力条件に従って、それぞれ転流パターン１〜４を生成して出力する。例えば転流パターン発生回路２１１に入力する転流パラメータＣａについて説明すると、次のようになる。パラメータＣａの現在値をＣａ（ｋ）、その１サンプル前のパラメータＣａをＣａ（ｋ−１）とする。

Ｃａ（ｋ）＝１のとき、 転流パターン１を発生
Ｃａ（ｋ）＝０でＣａ（ｋ−１）＝−１のとき 転流パターン２を発生
Ｃａ（ｋ）＝０でＣａ（ｋ−１）＝１のとき 転流パターン３を発生
Ｃａ（ｋ）＝−１のとき 転流パターン４を発生

転流パターン発生回路２１２及び２１３についても同様であり、Ｃｂ（ｋ）及びＣｂ（ｋ−１）、Ｃｃ（ｋ）及びＣｃ（ｋ−１）に基づいてそれぞれ転流パターン１〜４を発生する。転流パターン発生回路２１１の構成は、転流パラメータＣａ（ｋ−１）と１サンプリング（Ｚ^−１）遅れた転流パラメータＣａ（ｋ）を比較手段で比較し、両者が同じであれば非転流状態と判断し、ある瞬間で両者の出力が相違すれば転流の開始と判断し、その条件に従って予め登録されている転流パターンデータを読出して出力する。転流パターン発生回路２１２及び２１３についても同様である。

The commutation parameters Ca, Cb, and Cc generated and output in the above relationship are input to the commutation pattern generation circuits 211, 212, and 213, respectively. The commutation pattern generation circuits 211, 212, and 213 generate and output commutation patterns 1 to 4, respectively, according to the input conditions of the commutation parameters Ca, Cb, and Cc. For example, the commutation parameter Ca input to the commutation pattern generation circuit 211 will be described as follows. The current value of the parameter Ca is Ca (k), and the parameter Ca one sample before is Ca (k-1).

When Ca (k) = 1, commutation pattern 1 is generated. When Ca (k) = 0 and Ca (k−1) = − 1, commutation pattern 2 is generated. When Ca (k) = 0, Ca (k− 1) When commutation pattern 3 is generated when Ca = 1 is generated, commutation pattern 4 is generated when Ca (k) =-1.

The same applies to the commutation pattern generation circuits 212 and 213, and the commutation patterns 1 to 4 are generated based on Cb (k) and Cb (k-1), Cc (k), and Cc (k-1), respectively. . The configuration of the commutation pattern generating circuit 211 compares the commutation parameter Ca (k−1) with the commutation parameter Ca (k) delayed by one sampling (Z ⁻¹ ) using a comparison unit. If it is determined that the commutation state is present and the outputs of the two are different at a certain moment, it is determined that commutation is started, and the commutation pattern data registered in advance is read and output according to the condition. The same applies to the commutation pattern generation circuits 212 and 213.

When the commutation time is, for example, 6 milliseconds, the commutation patterns 1 to 4 are calculated and output with the numerical values shown in Table 3 below. Here, the maximum value of the commutation patterns 1 to 4 is “1”, and the minimum value is “−1”, and the decimal value is a decimal in the meantime. For this reason, in consideration of accuracy, 2 ¹⁰ is multiplied to become an integer. Therefore, in the graph of the commutation pattern (FIG. 4), numerical values that are “−1, −0.998296,..., 0,. -1024, -1007, ..., 0, ..., 1007, 1024 ". Thus, since it has been expanded by 2 ¹⁰ in advance, in order to eliminate the 2 ¹⁰ factor in advance, the output is multiplied by 2 ⁻¹⁰ as a coefficient.

なお、転流パターン１〜４の発生はホールセンサ信号が変化した場合、転流中でもいつでも受付けるようになっている。転流中に新たな転流を受付けないようにすると、ソフトウェアとして構造が複雑になり、ソフトウェアとしてのロバスト性が低下するからである。ホールセンサ信号が変化したということは、モータの位置パターンが変化したということであり、電流はモータ位置パターンにより流さないと、モータの制御ができないことになる。

The generation of commutation patterns 1 to 4 is accepted anytime during commutation when the Hall sensor signal changes. This is because if the new commutation is not accepted during the commutation, the structure of the software becomes complicated, and the robustness of the software decreases. The change of the Hall sensor signal means that the position pattern of the motor has changed, and the motor cannot be controlled unless the current is passed through the motor position pattern.

  The commutation patterns output from the commutation pattern generation circuits 211 to 213 are multiplied by multiplication circuits 221 to 223, respectively, and output as current command values Irefas, Irefbs, and Irefcs through coefficient circuits 231 to 233, respectively. The commutation waveform is generally shown in FIG. 5C, but in the present invention, the one phase is as shown in FIG. 5A and the three phases are as shown in FIG. 5B. .

In the present invention, a commutation pattern generation unit that generates a commutation pattern at the time of commutation is provided, and a current command value at the time of commutation is determined in advance based on the commutation pattern, that is, 2 of the commutation pattern. The current command value for each phase of the brushless motor is calculated by multiplying the commutation pattern and the current command value by using a set of numerical values where the sum of the phases is 1 or -1. For example, in the case of the commutation point 12, the commutation time is 12 × sampling time, and if it is divided into 12 equal parts by a sine wave waveform of ¼ period during that time, as shown in FIG.

[0.017037, 0.066987, 0.14645, 0.25, 0.37059, 0.5, 0.62941, 0.75, 0.85355, 0.93301, 0.98296, 1]
= [17, 69, 150, 256, 379, 512, 645, 768, 874, 955, 1007, 1024]

It becomes. And it is set as the numerical value set from which the sum of the two phases of a commutation pattern becomes 1 (1024 in an integer) or -1 (-1024 in an integer). Since the sum of the three-phase currents is 0 even during commutation, the sum of the commutation phase currents does not change during commutation so that the current of the non-commutation phase does not change even if the current of the commutation phase changes. It must always be the same size as the non-commutated phase and the direction reversed. In addition, if it is not restricted to a sine wave waveform and the sum of the two phases of a commutation pattern will be set to 1 or -1, it can commutate smoothly, without generating a torque ripple.

FIG. 6 shows an example of a linear waveform in which the sum of the two phases of the commutation pattern is 1 or −1. The commutation pattern 1 is a linear waveform of “0 → 1”, and the commutation pattern 2 is “− The linear waveform is 1 → 0, the commutation pattern 3 is a linear waveform of “1 → 0”, and the commutation pattern 4 is a linear waveform of “0 → −1”. FIG. 7 shows an example of a power function (1- ^{et / T} ) waveform in which the sum of the two phases of the commutation pattern should be 1 or −1. The commutation pattern 1 is “0 → 1”. It is a power function waveform, the commutation pattern 2 is a power function waveform of “−1 → 0”, the commutation pattern 3 is a power function waveform of “1 → 0”, and the commutation pattern 4 is “0 → −”. This is a power function waveform of 1 ″. It is also possible to use such a commutation pattern waveform.

  In the above-described embodiment, an example in which the present invention is applied to a three-phase brushless motor has been described. However, the present invention is not limited to this, and the brushless motor of other plural excitation phases (for example, five phases) can also be used. The present invention can be applied. In the above-described embodiment, as shown in FIG. 8, the example in which the upper transistor (for example, Ta1) and the lower transistor (for example, Ta2) are driven with the same PMW duty is shown. The present invention is not limited to this, and the present invention can also be applied to the case where the upper transistor (for example, Ta1) and the lower transistor (for example, Ta2) are driven with different PMW duties.

  According to the present invention, in order to suppress generation of vibration and noise during commutation for a brushless motor, a current command value is given as a sine wave waveform (or pseudo sine wave waveform) during commutation, and two phases of the commutation pattern are provided. Is given as a pair of numerical values whose sum is 1 (1024 for integer) or -1 (-1024 for integer), and the commutation pattern and current command value are multiplied to calculate the current command value for each phase of the brushless motor Like to do. As a result, even when the brushless motor is controlled by a rectangular wave, low vibration and low noise can be realized, and it can be applied to high performance electric power steering of automobiles and vehicles.

It is a circuit diagram which shows an example (example of a 5-phase brushless motor) of the motor drive control apparatus which concerns on the Example of this invention. It is sectional drawing of the 5-phase brushless motor which can apply this invention. It is a block diagram which shows the structural example (3 phase brushless motor) of the commutation pattern generation | occurrence | production part of this invention. It is a wave form diagram which shows the example of a commutation pattern (sine wave waveform). It is a wave form diagram which shows the example of the commutation waveform in a three-phase brushless motor. It is a wave form diagram which shows the example of a commutation pattern (linear waveform). It is a wave form diagram which shows the example of a commutation pattern (power function waveform). It is a wave form diagram which shows the example of the excitation electric current of each phase of a 5-phase brushless motor. It is a wave form diagram which shows an example of the electric current change of each phase at the time of the conventional excitation current switching, and an electromagnetic torque change.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Brushless motor 2 Housing 3a, 3b Bearing 4 Rotating shaft 5 Permanent magnet 6 Stator 7 Rotor 11 Phase detection element (Hall sensor)
DESCRIPTION OF SYMBOLS 20 Motor drive control apparatus 21 Control circuit 22 FET gate drive circuit 23 Motor drive circuit 24 Current detection circuit 25 Rotor position detection circuit 27 External circuit 211-213 Commutation pattern generation circuit 221-223 Multiplication circuit 231-233 Coefficient circuit

Claims (3)

In a drive control device of a brushless motor having a plurality of excitation phases driven by a rectangular wave excitation current, a current command value calculation means for calculating a current command value to the brushless motor, and a numerical value predetermined at the time of commutation A commutation pattern generator for generating a commutation pattern of each phase to be given, and each phase current command for calculating a current command value to each phase of the brushless motor based on the commutation pattern and a current command value A drive control device for a brushless motor, comprising a value calculating means. 2. The current command value for each phase is calculated by multiplying the commutation pattern and the current command value by a set of numerical values in which the sum of the two phases of the commutation pattern is 1 or −1. The brushless motor drive control device described. The brushless motor drive control device according to claim 2, wherein the commutation pattern has a sinusoidal waveform.

Images