TWI294717B - Motor driver, motor controller and controlling method for electric motor - Google Patents

Description

1294717 22134twfl .doc/006 95-11-20 IX. Description of the invention: [Technical field of the invention] The present invention relates to a motor drive technology, especially to motor control. A current vector control pulse width modulation inverter Device. [Prior Art] % Permanent magnet AC (AC) motor (PMACM) has been widely used in high-performance servos because of the following characteristics: high efficiency compared with direct current (DC) Lu motor. High torque to inertia ratio, low maintenance cost and streamlined construction. Permanent magnets are used to create substantially air gap flux without magnetic excitation, allowing them to design permanent magnet AC (AC) motors with excellent efficiency characteristics. This efficiency advantage has become increasingly valuable in many applications around the world. Because all permanent magnet AC (AC) motors are synchronous machines, the average torque can only be generated when the magnetic energy is accurately synchronized with the rotor speed and instantaneous position. The most straightforward and powerful way to ensure synchronization is to continuously measure the absolute angular position of the rotor with an installed position sensor (such as a Hall effect sensor). The magnetic (AC) motor phase synchronously switches the magnetic flux. One conventional method of achieving synchronization is to use a six-step six step voltage inverter. The basic operation of a six-step square wave voltage inverter can be understood by considering the inverter as six ideal switches. The line-to-line and line-to-line voltage and phase and line voltages have the waveforms shown in Figure 1. The line-to-line voltage contains a root mean square component, such as: ^ ll{rms fund) TVcc (1) Please refer to J. Holtz's "Pulse Width Modulation Survey", IEEE Trans. 1294717 95-11-20 22134twfl .doc/006

Ind. Electron·, Vol. 39, No. 5 ’Dan 410-420 pp.’ December 1992. In this document, a pulse width modulation (PWM) inverter maintains a nearly constant DC link voltage and includes voltage control and frequency control in the inverter. In operation, the power-on relationship in the inverter is switched at high frequency, so the actual operation can be regarded as a cut-off. In general, modulation techniques fall into two categories: operators with a fixed switching ratio at the basic switching frequency, and a switching ratio that continuously changes to synchronize a motor current closer to a sinusoid (called sinusoidal PWM). . In the first category, block modulation is the simplest of modulation and is closest to a simple six-step square wave operation. Different from changing the amplitude of the motor voltage waveform by changing the DC voltage, it is tied to a fixed switching ratio by switching one or both of the inverter switches.

Suitable for this speed. Figure 2 shows a simple form of block modulation, which is limited to the middle 6 〇 electrical angle of each component conduction period, resulting in a minimum switching period for the half-body switch. Regardless of the similarity between the block modulation mode and the six-step square-wave mode, the torque ripple (t〇rqu pulsatums) in the low-speed state is far less severe than the six-step inverter. In the absence of six steps, the situation of the county is also present in the block, but there is still a truncation frequency of the block modulation mode of higher spectral waves. Therefore, compared with the simpler Pantu::, the motor loss and noise of the module are quite significant. : angle;, two: ί voltage and current waveform. Even if the switches ΤΑ+ and ΤΑ- are at 18 (power factor, etc. d:), the conduction interval due to the delay of the load is less than the _ electrical angle. Current, two-string pulse width modulation (SPWM), which is used to synthesize the motor as close as possible to the sinusoidal waveform. This (4) method can greatly attenuate the voltage chopping of the low 1294717 22134twfl.doc/006 95-11-20, usually leaving only the second harmonic or the fourth harmonic whose amplitude is close to the truncation or wave frequency. Compared with the six-step square wave motor, the motor can rotate more smoothly at a lower speed and substantially eliminate torque ripple, and the extra motor loss generated by the inverter varies with sinusoidal pulse width modulation (SPWM) operation. The efficiency is substantially reduced. However, to balance this advantage, sinusoidal pulse width modulation (SPWM) inverter control is more complex, 'and the cutoff frequency is higher, resulting in higher switching losses than the six-step square wave operation. To approximate a sine wave, a comparison of a high frequency triangular wave with a fundamental frequency sine wave is shown in Figure 4. Current control technology plays the most important role in current-controlled pulse width modulation (PWM) inverters, which have been widely used in south performance motor drives. The various techniques of the current controller have been described in the following documents [1]-[7]: [1] M. Lajoie-Mazenc, C. Villanueva and J. Hector "The Hysteresis Control Inverter on Permanent Magnet Synchronous Machines Study and implementation of hysteresis controlled _ inverter on a permanent magnet synchronous machine·, IEEE Trans. Ind. Applicat, IA-21, No. 2, pp. 408-413, March 1985/ April. [2] D. M. Brod and D. W. Novotny, "Current control of VSI-PWM inverters", IEEE Trans·Ind. Applicat, Volume IA-21, No. 3, pp. 562-570, May/June 1985. [3] Τ·M. Rowan and R.J. Kerkman's analysis of novel synchronous current regulators and current-regulated PWM inverters (A new 7 1294717 22134twfl.doc/006 95-11-20 synchronous current regulator and an Analysis of current-regulated PWM inverters·)", IEEE Trans. Ind. Applicat, Volume IA-22, No. 4, pp. 678-690, July/August 1986. [4] M. P. Kazmierkowski, MA Dzieniakowski and W. Sulkowski, "Novel space vector based current controllers for PWM-inverters", IEEE Trans. Power Electron, Vol. 6, No. 1, pp. 158-166, January 1991. [5] CT · Pan and Τ·Υ· Chang "An improved hysteresis current controller for reducing switching frequency" IEEE Trans Power Electron, Volume 9, No. 1, Pages 97-104, 1994. [6] L. Malesani and R Tenti's "A novel hysteresis control method for current-controlled voltage-source PWM with a fixed frequency modulation current-controlled voltage source pulse width modulation (PWM) inverter Inverters with constant modulation frequency)", IEEE Trans. Ind. Applicat, Vol. 26, No. 1, pp. 88-92, January/February 1990. [7] S. Buso, S. Fasolo, L. Malesani, and P. Mattavelli, “A type of adaptive hysteresis current control (A dead-beat 1294717 95-11-20 22134twfl.doc/006 adaptive hysteresis current control· ), IEEE Trans. Ind. Applicat, Vol. 36, No. 4, pp. 1174-1180, July/August 2000. However, in the above-mentioned prior art, the hysteresis current controller (HCC) is quite popular because of its easy implementation, fast dynamic response, maximum current limit, and insensitivity to load parameter degeneration. However, depending on the load conditions, the switching frequency may vary greatly in amplitude during the basic period, resulting in irregular inverse and cross-talking characteristics. This is mainly due to mutual interference between the three phase transitions, since each phase current depends not only on the corresponding phase voltage but also on the voltage of the other two phases. Therefore, the actual current waveform is not only determined by the hysteresis control, but also depends on the operating conditions. The current slope may vary considerably, and the current peak may significantly exceed the = rate of the hysteresis band, which is much higher than the demand to meet the continuous wave and the miscellaneous 4;:: = (4) must be adjusted accordingly. In addition, high frequency and scale values are added and may affect system reliability. Some techniques for applying zero vectors to reduce the lag and lag current control techniques have recently been proposed [4]·[5]. In addition, the household level of the customary method: the interference effect is minimized, and the hysteresis azimuth locking loop is maintained (Ρ·2:1—the phase of the Yunmei amplitude [q-mn k has a fixed switching frequency in the cycle; Single) The control of the function ^ %L aspect 'space vector modulation (SVM) _ with two excellent 匕 匕, eight at the same carrier frequency 1294717 22134twfl .doc / 006 95-11-20 maximum output voltage is 15.4% high, and The number of switches is about 30% less. [8] K. Zhou and D. Wang, "Relationship between space-vector modulation and three-phase carrier-based PWM * A comprehensive analysis", IEEE Trans·Ind· Electron·, Vol. 49, No. 1, pp. 186-196, February 2002. [9] V.Blasko's "Modified Space Vector and Triangle Comparison Method"

Analysis of a hybrid PWM based on modified space-vector and triangle-comparison method, IEEE Trans·Ind. Applicat, Vol. 33, pp. 756-764, May/6, 1997 month. [10] X. Xu and D. Deng, "Three phase inverter circuit with improved transition from SVPWM to six step operation", U.S. Patent No. 5,552,977, issued to Ford Motor Company, September 3, 1996. [11] V. Blasko, "Mixed Pulse Width Modulation Method and Apparatus", US Patent No. 5,706,186, Allen-Bradley, January 6, 1998. [12] Β·H· Kwon\T·W· Kim and J. H. Youm, "A novel SVM-based hysteresis current controller", IEEE Trans Power Electron, Vol. 13, No. 2, pp. 297-307, March 1998 . 1294717 22134twfl .doc/006 95-11-20 The SVM technique is used to limit the space vector to be applied based on the region of the output voltage vector. However, to achieve zero output current error, SVm technology requires an unrealistic back EMF vector measurement. HCC allows the output current vector to follow the command vector with almost negligible response time, line voltage, and insensitive load factor changes. However, in addition to the required space vector, HCC also generates other vectors depending on the region in the SVM technique. If a zero vector is applied to reduce the current value of the output current vector, the line current will decrease with a gentle slope and reduce the switching frequency. SVM-HCC using all the features of HCC and SVM technology has been exposed in [12]. To control the three-phase current of the motor, an effective method is directly measured by three low-value resistors or a Hall effect current sensor. However, this method is not economical. If the motor windings are star-connected, the number of sensors driven by the three-phase motor can be reduced to two. However, this method introduces an error in the estimation of the third phase current because of the difference in gain constants and the DC offset of the other current sensors. An alternative approach is to reconstruct the three phase currents based on the measured DC link current and pulse width modulation (PWM) signals, as described in the following literature and patents [13]-[21].

[13] O·P· Acarnley's "Observability criteria for winding currents in three-phase brushless DC drives", IEEE Trans. Power Electro·, Volume 8 'No. 3', pp. 264-270, July 1993. [14] C·D·French, P·P·Acarnley and A·G· Jack in a brushless driver using a single DC-connected current sensor 11 1294717 22134twfl.doc/006 95-11-20 Instant current Real-time current estimation in brushless DC drives using a single DC-link current sensor·), EPE Conf· Rec., 1993, pp. 445-450. [15] J. F. Moynihan, S. Bolognani, RC Kavanagh, Μ·G·Egan, and J.M. D. Murphy “Single sensor current control for AC servo drives using digital signal processors (Single sensor) Current control of AC servo drives using digital signal processors·)”, EPE Conf· Rec·, 1993 _, pp. 415-421. [16] J. Zhang and M. Schroff, "Current control of three-phase brushless DC drives with DC-link current measurement", Power Conv. Intell. Motion (PCIM) Conf·, pp. 141-148, June 1997.

[17] F. Blaabjerg, JK Pedersen, IL Jaeger and R _ Thoegersen "Single current sensor technique in the DC link of a three-phase PWM-VS inverter: a single current sensor technique in the DC link of three-phase PWM-VS inverters: A review and a novel solution·)", IEEE Trans. Ind. Applicat, Vol. 33, No. 5, pp. 1241-1253, September/October 1997.

[18] H. Tan and S. L. Ho, "A novel single current sensor teaching suitable for BLDCM drives", IEEE-PEDS <:s > 12 1294717 22134twfl.doc/006 95-11-20

Conf·, 1999, pp. 133-138. [19] L. Ying and Ν Ertugrul, "A novel estimation of phase currents from DC link for permanent magnet AC motors.", Conf· Rec·, Pp. 606-612, 2001. [20] improved· Μ · Wolbank and R Macheiner "An improved observer-based current controller for inverter fed AC machines with single DC_link current Measurement·)", IEEE_PESC Conf. in Proc., 2002, pp. 1003-1008. [21] P·Yu's "Phase current sensor using inverter leg shunt resistor", Texas Instruments, US Patent No. 3, 2003 6,529,393. According to the concept of SVM, the inverters feeding the motor have only eight possible switching patterns, which are represented by two kinds of zero-sorrow and six action states. During the six states of operation, only one of the three phase currents flows through the DC link. However, in the two zero states, the phase current circulates through the diode in the inverter bridge without DC connection. In the pulse width modulation (PWM) current control mode, there are two possible states of action in each modulation cycle. Therefore, the two-phase current can be derived from the DC link current. However, under certain operating conditions of pulse width modulation (pWM) control, either of the two modes of action may last for a very short period of time. Therefore, due to the limited switching time of the power components, the dead time of the electronic components, and the delay in the electronic circuit, the actual phase current may not be measured on the DC connection measurement. Figure 5 is a block diagram of a conventional six-step square wave motor driver, wherein the motor driver includes A phase, B phase, and C phase upper side driving transistors 101, 103, and 105, U phase, V phase, and W phase. Side drive transistors 102, 104 and 106, diodes 101D, 102D, 103D, 104D, 105D and 106D, a Hall sensor circuit 201, a conventional six-step square wave control circuit 202, a front The driving circuit 203 and a current price measuring resistor 204 are provided. A motor includes an A phase coil 301, a B phase coil 302, and a C phase coil 303. In this embodiment, an N-type metal oxide semiconductor (nm〇S) electro-crystalline system is used to drive the transistors 101-106. The anode terminal and the cathode terminal of the diode i〇id are respectively connected to the source terminal and the drain terminal of the driving transistor 101. Similarly, the anode and cathode ends of the diodes 102D-106D are connected to the source terminals and the gate terminals of the drive transistors 102-106, respectively, in the same manner. The drain terminals of the driving transistors 10, 103 and 105 are connected to the power supply Vcc, and the source terminals of the driving transistors 102, 104 and 1 are connected to one end of the current detecting resistor 204. The other ends of the current detecting resistor 2〇4 are grounded. The diode 1〇1D_i〇2D of the driving transistor 101_102 operates as an A phase output circuit, and the diode 103D-104D of the driving transistor 1〇3_ι〇4 operates as a B phase output circuit and drives the transistor. The arm and the diode face of the 105-106 are operated as a c-phase output circuit. The source terminal of the transistor ι〇1 and the common terminal of the transistor of the transistor 1〇2 are connected to the A-phase coil 3. 〇1's terminal. Similarly, the common terminal of the source terminal of the crystal 103 and the drain terminal of the transistor i〇4 is connected to a terminal of the B phase coil 302, and the electric terminal 14 < £ 1294717 22134 twfl. doc / 006 95-11-20 A common terminal of the source terminal of the transistor 1〇5 and the drain terminal of the transistor 106 is connected to a terminal of the c-phase coil 303. The A phase coils 3〇1, b phase coil 302, and the other ends of the C phase coil 303 are connected to each other. The current flowing from the driving transistor 10M02 to the A-phase coil 301 is referred to as the A-phase current IA. Similarly, the current flowing from the driving transistor i〇3-1〇4 to the B phase coil 302 is referred to as the B phase current IB, and the current flowing from the driving transistor® 105-106 to the C phase coil 303 is referred to as the C phase current. 1C. The direction of all phase currents ΙΑ, IB, and 1C flowing from the driving transistor 10M06 to the coils 301-303 is assumed to be a positive direction for all phase current systems. The coils 301-303 of the motor 300 are connected in Y. Therefore, each phase current is equal to the current flowing through the corresponding coil. The Hall sensor circuit 201 includes Hall sensors 201A, 201B, and 201C that detect the position of the motor 300 rotor and output detection results to the position detection circuit and current command generation. Circuits 22, such as Hall sensors 201A, 201B, and 201C, output H1+, Η1-, Η2+, Η2-, Η3+, and H3_. A conventional six-step square wave control circuit 202 (which receives Hall sensor outputs H1+, H1_, H2+, H2-, H3+, and H3-, a torque command signal Tc, and a feedback current signal Ifb) The control signals S11-S16 are switched to select to turn any of the drive transistors 101-106 on or off and to transmit an instruction to the front drive circuit 203. The front drive circuit 203 outputs a signal to the gate of the drive transistor 101-106 according to the output of the conventional six-step square wave control circuit 202 to control the drive transistor 101-106 15 1294717 22134 twfl .doc/006 95-11-20 On/off status. Figure 6 is a block diagram of a conventional six-step square wave control circuit including differential amplifiers 401A, 401B, and 401C, automatic gain control circuits 402A, 402B, and 402C, adders 403A, 403B, and 403C. Multipliers 404A, 404B and 404C, comparators 405A, 405B, 405C, 412A, 412B and 412C, a low pass filter 406A, a peak detection circuit 407, an adder 408, a controller 409, a carrier signal generation The device 410 and an invalid time control circuit 411. The differential amplifiers 4〇1 a, 401B, and 401C receive Hall sensor outputs H1+, H1_, Η2+, Η2·, Η3+, and H3-, respectively, according to Hall sensor outputs H1+, HI-, H2+, H2-, H3+, and H3- determine position signals Ha, Hb, and He, and output position signals Ha, Hb, and He to automatic gain circuits 402A, 402B, and 4020. Automatic gain circuits 4〇2A, 402B, and 402C adjust position signals Ha. The current values of Hb and He then generate signals H11, H21 and H31. After receiving the signals ml, h21, and H31, the adder generates signals H13, H23, and H33 to comparators 412A, 412B, and 412C, respectively. The low pass filter 406, which receives a current feedback signal Ifb, outputs a signal to the peak debt measuring circuit 407. The adder 4〇8 receives the detection result generated by the torque command signal tc and the peak 4 detection circuit 407, and outputs the error signal to the controller 4〇9° to receive the output signal of the controller 4〇9 and the comparator 412a, respectively. The multipliers 4〇4a, 404B and 404C of the output signals of 412B and 412C will output the results to the comparators 4〇5A, 4〇5b and 405C. The invalid time control circuit 411 determines the switching control signals S11-S16 based on the outputs of the comparators 405A, 405B, and 405C. 16 1294717 22134twfl .doc/006 95-11-20 Figure 6 shows the control block diagram of the current control architecture of the spindle motor. The basis of this control layout is similar to open loop voltage/frequency control. The amplitude and phase of the voltage are individually controlled. This control layout has two limitations. Since the DC link current is dependent on the Pulse Width Modulation (pWM) signal, a stop current is measured as current feedback as shown in Figure 7. After detecting the peak value of the Dc junction current, a continuous current feedback can be generated as shown in Fig. 7. However, the current feedback generated by it contains large ripples, which may cause poor current control performance even in steady state operation. In addition, the control parameters of the current controller 409 shown in Fig. 6 need to be adjusted to improve the control efficiency when applied to different motors. Figure 8 shows the simulation results of the current control performance using the conventional six-step square wave control architecture. Figure 9 shows the control block diagram of the modified six-step square wave control circuit. The three comparators 412A, 412B, and 412C are omitted from FIG. Figure 10 shows the current control performance simulation waveform of the modified six-step square wave control architecture. Since only the amplitude of the maximum phase current is controlled, the controlled phase current is similar to the trapezoidal waveform shown in Fig. 10. In addition, since the torque generated by the non-sinusoidal phase current contains a torque ripple, it may cause the motor to oscillate and may deteriorate the efficiency. In the conventional method, whether it is block modulation or sinusoidal pulse width modulation (SPWM), there is a problem that only the amplitude of the maximum current can be controlled. Therefore, the shape of the phase current cannot be controlled. In U.S. Patent No. 6,674,258, Matsushita Electric Industrial Co., Ltd. has proposed a current control architecture capable of controlling two phase currents in a pulse width modulation (PWM) switching cycle. Figure 11 shows the overall control block diagram of the Matsushita method. For simplicity, 17 1294717 22134twfl.doc/006 95-11-20 Two trapezoidal current commands are generated as shown in Figure 12. The time interval TU1 in Fig. 12 is taken as an example to explain the basic principle of this control method. During this time interval, the terminal voltage for phase a is limited to Vcc as shown in Fig. 13(a), and the phase current ia is controlled according to the torque current command TI. Since only one phase current can be sensed from the DC link current, the other two terminal voltages are switched to ground for the start of the pulse width modulation (PWM) switching cycle as shown in Figure 13(a). Phase Lu potential current ia. When ia reaches the torque current command, the lower switch of phase b is turned off by the control signal F1, and the phase current ib flows through the diode 3D of the upper switch as shown in the figure 。. After F1 is turned off, a negative value of the phase current ie can be sensed from the DC link current, and it is controlled to follow the ramp current command TP ' as shown in Fig. 14(a) - 14(b). When the negative ic reaches the ramp current command TP, the lower switch of phase 3 is turned off by the control signal F2, and the phase current ic flows through the diode 5D of the upper switch as shown in Fig. 13(c). In theory, this method not only controls the amplitude of the maximum phase current, but also controls the shape of one of the other two phase currents during a pulse width modulation (PWM) switching period. Figure 15 shows the simulation results of the Matsushita method. From this figure, due to the non-sinusoidal current waveforms, the generated torque contains a large torque ripple. It should be noted that the controlled phase current is not the ideal trapezoidal waveform required in Figure 12. The reason will be explained below. In fact, this method has a fundamental problem for controlling two phase currents in a pulse width modulation (PWM) switching period. Again, the time interval TU1 in Fig. 12 is taken as an example. At the beginning of the pulse width modulation (Pwm) switching period, the phase current ia is controlled toward the torque current command TI. However, 1294717 22134twfl.doc/006 95-11-20

The negative value of the phase current ic is also increased as shown in Fig. 16. When the phase current ia reaches the command demand, the phase current ic may have exceeded the ramp current command as shown in Figure 16. Therefore, the shape of the phase current ic cannot be controlled until the ramp current command exceeds the negative value of the phase current ic. Figure 17 (a) shows that even if the current command is a three-sine waveform, this basic problem can still occur. Another observation in Fig. 17(b) is that if the shape of the controllable current is ib instead of ic, the phase current ib can be controlled until the current command is lower than the negative value of the phase current ib. Therefore, a reasonable answer to this basic problem is to control ia and ib in the first half of TU1 and ia and ic in the second half of TU1. Mathematical analysis will be provided in the following sections to explain the basic issues of this method. From the 13th (a) diagram, the following three-phase voltage equation can be derived:

Van = Va~Vn = + 0 + °) = + + (^) vbn =vb^vn =^-~(Kc + 〇 + 〇) = ibR + L~^ + eb (3)

Vcn =Vc -Vn =Q--(Kc+^ + Q) = hR + L-^ + ec (^) where Van, Vbn, Vcn are three-phase voltages, Va, Vb, Vc are three terminal voltages, and vcc is DC Connect the supply voltage, ia, ib, ic are three-phase current, ea, eb, ec, three post-electromotive voltage, R and L are stator resistance and inductance. From the above equation, the change in phase current can be estimated as:

Aial=j(^Vcc^ea-iaR] (5) ^=\[-\vcc-eb-ibR\ (6)

L\ 0 J ^=\[-\vcc-ec-icRj (7) A similar analysis can be performed for Figure 13(b), which becomes: K2=\i\Kc-ea-iaR\ (8)

Aib2=y -vcc~eb-hR (9) 19 1294717 22134twfl.doc/006 95-11-20 . 1 ( 2 > Alc2 =7[~3FcC_eC""/ci? (10) ) Pictured: Aia 〇=j{-ea-^iaR) (11) 1 ^h〇=j(-eb-ibR) (12) Aic〇=J^{-ec-icR) (13) -r蚵弟u(8) to (8) The ambiguous time intervals are 岣, A, and Δ&, where η refers to the ηth switching period in the time interval TU1. It can be deduced that the phase current ic at the moment of the change is as follows: k (14) From (7), (10) and (13), (14) can be derived as: -1 moxibustion _ 1 (15) Σ /d + 2Δ, „2)+〜Δ7Ά ^ n=\LJ nx sm where ... the switching period 'is also the sum of <, ^ and ^. With the trapezoidal current waveform shown in Figure i2, the kth switching instant can be derived Phase current command, ·:: Let A (16) have a magnetic pole in which the rotary shaft motor is indicated by P, respectively, and ω means the rotational speed at the first switching instant ' and r refers to the amplitude of the current command. As mentioned above, If you want to control the phase current ic to cover the current in a time interval of τυι, hk — hk will be rewritten as: (17)

Ko △Κ (18) The result equation (18) holds ' _ ^ ' at the kth switching instant to control the phase M ieH TU1 _ k switching instant followed by current life 20 1294717 95-11-20 22134twfl.doc/006 order, ·:. Therefore, equation (18) is a condition that determines whether the shape of the phase current ie is controllable. Some of the following observations can be obtained from equation (18). Within the TU1 time interval, the first term on the left side of equation (18) is a positive value, and the second term is a negative value, ie: 〇&lt;-Kc (19) ecn + icnR< Q (2 〇) The right side of (18) is the rotation speed ω. And a wide range of phase currents

Directly proportional. Therefore, as shown in Figs. 18(a)-18(b), equation (18) is more easily satisfied at a higher speed than at a lower speed. It is assumed that the back electromotive voltage and the phase current are both sinusoidal and can be derived as follows at the time interval: /, ecn = -ΚΕωη sin 20 -ΚΕωη sin(3Pco0«A7^) 4 =^Ψ^ΪΏ{3Ρω0ηΑΤ^) (21 (22) At low speeds, the right side of equation (18) is approximately zero. So for the negative summation to the left of equation (18), you get:

Fcco" + , )sin(3 corpse ω0πΔ7^ ) prepared by Fcc 1 + △’2

AT (23) From the condition of equation (23), a knot l can be obtained, that is, equation (23) can only be satisfied when operating at low speed and having a sufficiently large η, that is, at low speed operation, the entire time interval of TU1 cannot be achieved. The shape of the phase current ic is controlled internally. Please refer to Figure 18(a)-18(b). This phenomenon may cause torque ripple to affect the overall control efficiency. From the above analysis, the concept of the Matsushita method has several advantages. First, not only can the amplitude be controlled, but the shape of the phase current can be controlled to reduce torque ripple. Second, there is no need to adjust any control parameters. Third, in the 21 1294717 22134twfl.doc/006 95-11-20, only two phases need to be switched, so the switching cost of the power transistor can be reduced. However, Matsushita's method has fundamental problems in controlling the shape of current. Therefore, a new control architecture was proposed to preserve the advantages of the Matsushita method and to improve its shortcomings. SUMMARY OF THE INVENTION The main purpose of the present invention is to provide a space vector current control pulse width = Ai (PWM) system motor drive technology, the drive technology can control a plurality of phase currents do not follow a DC link current feedback And changed strongly. Another object of the invention is to provide a motor driver based on space vector current control pulse-modulated dual (PWM) technology. In the motor driver, the amplitude of the phase current and its shape can be controlled to reduce torque ripple. Another object of the present invention is to provide a motor driver based on space vector current controlled pulse width (PWM) technology. In this motor drive, only two phases need to be switched at any instant. Therefore, the switching loss of the power transistor can be reduced. Therefore, in order to achieve the above object, the present invention provides a motor driver for driving a motor. The motor includes a plurality of stator coils, and one end of each stator coil is commonly connected to a common node, and the motor driver includes a plurality of An output circuit, each output circuit includes an upper arm switch and a lower arm switch, wherein the connection between the upper arm switch and the lower arm switch of the i-th output circuit is connected to the i-th stator coil; the current sense resistor is switched The plurality of output circuits; the position detecting circuit is configured to output a ~ position signal related to the position of the motor rotor; and the current command generating circuit is configured to generate a plurality of predetermined current command signals according to the position signal and the predetermined phase angle to correspond to the stator coil · 22 &lt;.s 1294717 22134twfl.doc/006 95-11-20 Space vector modulation control circuit, receiving the above predetermined current command signal and engaging current 2 resistance, when the phase k of the kth predetermined current command signal When the positive and negative preset phases are controlled, the upper arm switch and the lower arm switch of the output circuit are controlled to receive the kth stator captured by the current detecting resistor Current loops, the current in the stator winding comparison of the k-th and the k th predetermined size ^ current command signal to select one from the stator coil, and to control the magnitude of the current u, wherein i and k is a natural number. ' _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ A preferred embodiment of the invention. It is to be understood that the invention may be embodied in various embodiments and the various details may be modified in various obvious aspects and all without departing from the invention. Accordingly, the drawings and description are to be regarded as illustrative rather [Embodiment] FIG. 19 is a block diagram of a motor driver according to a preferred embodiment of the present invention, wherein the motor driver includes a Hall sensor circuit 501 and a position detector. Measuring circuit and current command generating circuit 502, a space vector modulation (SVM) control circuit 503, a pre-drive circuit 504, a current detecting resistor 505, a U-phase, a V-phase, and a W-phase upper driving transistor 601, 603 and 605, U-phase, V-phase, and W-phase lower side drive transistors 602, 604, and 606, and complex diodes 601D, 602D, 603D, 604D, 605D, and 606D. A motor includes a U-phase coil 701, a V-phase coil 702, and a W-phase coil 23 1294717 22134 twfl .doc/006 95-11-20 703 In this embodiment, an N-type metal oxide semiconductor (NM〇s) A crystal system is used to drive the transistors 601-606. The anode terminal and the cathode terminal of the diode 601D are respectively connected to the source terminal of the driving transistor 601 and the drain terminal. Similarly, the anode terminal and the cathode terminal of the diode 002D_606D are respectively connected to the source terminal and the drain terminal of the driving transistor 6〇2_6〇6 in the same manner. The driving terminals of the driving transistors 6〇6〇3 and 6〇5 are connected to the power supply Vcc, and the source terminals of the driving transistors 6〇2, 6〇4 and 6〇6 are connected to the current. One end of the resistor 5〇5 is detected. The other ends of the current detection resistor 505 are grounded. The arm of the driving transistor 6〇1_6〇2 and the diode 601D-602D operate as a u-phase output circuit, and the arm of the driving transistor 603-604 and the diode 603D_604D operate as a v-phase output circuit. The arms of the drive transistors 605-606 and the diodes 605D-606D operate as a w-phase output circuit. The common terminal of the source terminal of the transistor and the drain terminal of the transistor 602 is connected to a terminal of the φ U phase coil 701. Similarly, the source terminal of the transistor 6〇3 and the common node of the drain terminal of the transistor 604 are connected to a terminal of the v-phase coil 702, and the source terminal of the transistor 6〇5 and the battery The common node of the drain terminal of crystal 606 is coupled to a terminal of the cargo phase coil 703. The U-phase coil 701, the V-phase coil 702, and the other terminals of the W-phase coil 703 are connected to each other. The current flowing from the driving transistors 601-602 to the U-phase coil 701 is referred to as U-phase current ;; Similarly, the current flowing from the driving transistor 6〇3_6〇4 to the V-phase coil 7〇2 is referred to as a v-phase current Iv, and from the 24 1294717 22134 twfl .doc/006 95-11-20 driving transistor 605- The current flowing to the W phase coil 703 is referred to as the W phase current Iw. The directions of all the phase currents Iu, 1¥, and lw flowing from the driving transistors 601-606 to the coils 701-703 are assumed to be positive directions for all phase current systems. The coils 701-703 of the motor 700 are Y-connected. Therefore, the respective phase currents are equal to the current flowing through the corresponding coils. The Hall sensor circuit 501 includes Hall sensors 501A, 501B, and 501C for detecting the position of the motor 700 rotor, and outputting the detection result to the position detecting circuit and the current command. A generating circuit 502, such as Hall sensors 501A, 501B, and 501C, outputs H1+, HI-, H2+, H2-, H3+, and H3-. The position detecting circuit and the current command generating circuit 502 determine the position signals Hu, 1^ and Hw according to the Hall sensor outputs H1+, H1-, H2+, H2_, H3+ and H3_, and output the position signal Hu, Hv and Hw to SVM control circuit 503. The position signals Hu, Hv and Hw are coefficient bit signals. The position detecting circuit and the electric current command generating circuit 502 are based on a torque command signal Tc, a required phase offset angle 0, and a Hall sensor output H1+, HI-, H2+, H2-, H3+. And H3-, determining the U phase current command signal /;;, the V phase current command signal /;, and the W phase current command signal 4. The position detecting circuit and the current command generating circuit 502 output a U phase current command signal &, a V phase current command signal &lt; and a W phase current command signal anvil to the svm control circuit 503. The SVM control circuit 503 receives the position signal Hu, the private and HW, the U phase current command signal 4, the V phase current command signal &lt; and the W phase current command signal 4 and a feedback current signal Ifb to generate a switching control.

&lt; S 25 1294717 22134 twfl .doc/006 95-11-20 Signals S21-S26 are selected to turn any of the drive transistors 6〇1-6〇6 on or off and transmit commands to the pre-driver circuit 504. The pre-driver circuit 504 rotates the signal to the drive transistor according to the output of the SVM control circuit 503.

„ The gate of body 601_606 to control the ON/OFF state of the drive transistor 601-606. Fig. 20 (a)-(b), Fig. 20(a) shows the space vector and the switch mode Figure (b) is a predetermined waveform for the respective phase currents according to a preferred embodiment of the present invention. The space vector modulation treats the driving transistors 601-606 in Figure 19 as a unit. The unit can be driven into eight unique states, each of which produces a different voltage vector. These states are shown in Figure 20(a), where the vector is represented by i, which refers to an upper side drive transistor (eg, The upper side driving transistor 6〇1, 6〇3 or 605) of Fig. 19 is turned on, and the lower side driving transistor (i.e., the lower side driving transistor 602, 604, or 606 of Fig. 19) is turned on. In the second diagram (a), the condition of a transistor off is indicated by a short line extending from an upper supply voltage or a lower supply voltage. Conversely, a transistor is turned on by extending downward and The longer line extending to the right (ie towards the stator winding) is indicated. The voltage vector V0' is for example All lower drive transistors are opened to short-circuit the stator windings. «Vector V7, also shorts the stator windings by opening all upper drive transistors. Therefore, the voltage vectors V〇 and V7 correspond to zero voltage in the stator windings due to their It is called a null or zero vector. - The electric dust vector VI transmits the current through the upper-side driving transistor to its respective stator winding, group, and then separates the current to pass the other two stator windings and the like. The respective lower side drive transistors. A voltage vector V2 from two 26 1294717 22134twfl.doc/006 95-11-20 upper side of the transistor allows current to pass through its respective stator windings, and then combines these currents into a current through Remaining stator windings and their respective underside transistors. From these examples, the switching states of other voltage vectors can be seen from examining Figure 20(a). Figure 20(a) shows eight switching states and indicates these The voltage vector of the state. Further, Fig. 20(b) shows a region in which the space vector according to the present invention is defined as a sinusoidal current command. In the 20th (b)th diagram, the voltage vectors are mapped to the axis of a state diagram. The The empty vectors V0 and V7 are positioned at the center of the coordinate, the voltage vector VI is placed along the α-axis, and the voltage vector V2-V6 is continuously separated by 60 degrees from the voltage vector VI. Therefore, the axis of the state diagram can be divided into six. It is to be noted that the present invention differs from the regional definition of the Matsushita method shown in U.S. Patent No. 6,674,258 and U.S. Patent Publication No. 2004/0000884, as shown in Fig. 20(b). Improving the weakness of the pattern tracking ability of the Matsushita method as described in the foregoing. ^ Referring to Figure 21 of the drawings, Figure 21 shows a position detecting circuit and a current command generating circuit according to a preferred embodiment of the present invention. schematic diagram. The position detecting circuit includes differential amplifiers 801U, 801V, and 801W, automatic gain control circuits 802U, 802V, and 802W, level shift circuits 803U, 803V, and 803W, and comparators 804U, 804V, and 804W. The position detecting circuit determines the position signals Hu, Hv, and Hw indicating the rotor position of the motor 700 based on the Hall sensor outputs H1+, Η1_, H2+, H2-, H3+, and H3-. The output of the differential amplifier 801U represents the difference between the Hall sensor outputs H1+ and H1-. Similarly, the 27 1294717 22134 twfl.doc/006 95-11-20 output of the differential amplifier 801V represents the difference between the Hall sensor outputs H2+ and H2-. The output of the differential amplifier 801W represents the difference between the Hall sensor outputs H3+ and H3-. The automatic gain control circuits 802U, 802V, and 802W receive the outputs of the differential amplifiers 801U, 801V, and 801W to adjust the outputs of the difference amplifiers to have the same peak value. Therefore, the outputs HH, H12, and H13 of the automatic gain control circuits 802U, 802V, and 802W have the same amplitude. Since the Hall sensor outputs H1+, H1-, φ H2+, H2_, H3+, and H3_ are approximately sinusoidal, the signals HH, H12, and H13 are also approximately sinusoidal. The phase of the signal H11 is 120 degrees ahead of the phase of the signal H12. Similarly, the phase of signal H12 is 120 degrees ahead of the signal of signal H13. a level shifting circuit 803U for shifting the voltage levels of the outputs H11, H12, and H13 of the automatic gain control circuits 8〇2U, 802V, and 802W,

The 803V and 803U' outputs the results to the comparators 8〇4U, 804V and 8〇4W, respectively. The comparators 804U, 804V, and 804W compare the outputs of the level shift circuits 803U, 803V, and 803W with a voltage reference #Vref, and generate position signals Hu, Hv &amp; Hw, respectively. The current command generating circuit includes multipliers 805a-805f, adders 8〇6U, 806V and 806|, multipliers 8〇7U, 8〇7V and 8〇7W, a phase shift table 808, and a torque amplitude ratio Gain control circuit 8〇9. The phase offset table 808 determines the values of &amp;&amp;&amp; based on the required phase offset angle 0. Position ^ then b tiger mi is from Κι*Η11_Κ2*Ηΐ2. Similarly, the location debt, 遽Η22 comes from the position signal Η23 from 1 ίί13_Κ2 H11. Suppose Κι=Κ2=1. Therefore, the phase of the positional debt signal book 28 1294717 22134twfl .doc/006 v 95-11-20 is 3 degrees ahead of the signal H11. In other words, the phase of the H21 advance signal Ή 11 μ position detection signal determines the phase offset angle θ from the values of the heart and κ. The same value, that is, the phase of the HU is required by K f, and the signal of the _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ The demand phase bias phase currents 1 and ; ^ are determined by the signal and the torque command signal TC. Torque ^^糸 hunting

The value is obtained by the torque vibration ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Determined with the rib and torque command signal Tc. As a result, the port phase current command signal /;, v ^ eye current command confidence and w phase current life signal /; according to the - torque command signal Tc, - demand phase offset angle 0 and Hall (Hall) sensor Outputs H1+, H1_, H2+, H2, H3+, and H3-. The phase shift table 8〇8 shown in Figs. 22 and 23 shows the output of the position detecting and current command generating circuit in accordance with a preferred embodiment of the present invention. Referring to Figures 24 through 26 of the drawings, Figure 24 shows a schematic diagram of an SVM control circuit in accordance with a preferred embodiment of the present invention. Figure 25 is a timing chart showing an SVM control circuit in accordance with a first preferred embodiment of the present invention. Figure 26 is a diagram showing a look-up table of an SVM control circuit in accordance with a first preferred embodiment of the present invention. The SVM control circuit 503 includes a multiplexer 901, an inverter circuit 902, a quasi-offset circuit 903, a low pass filter 904, a quasi-offset and amplifier 905, a comparator 906, and a space. Vector modulation 907, a reference clock generator 9〇8, 1) type flip-flop cs 29 1294717 22134twfl.doc/006 95 11 20 909 and 911, a delay 910, a falling edge delay 912, the inverter 9i3 And 916, NAND gate 914, and a lookup table 917. The lookup table 917 determines the conduction state of the multiplexer 901 and the output of the space vector modulation 907 based on the position signals Hu, Hv, and Hw, the detection state signal DS, the control state signal CS, and the state signal SS. The lookup table 917 also determines the state of the inverter circuit 902. For example, suppose SS = 〇, DS = 1, CS = = 〇, η = 1, Ην = 0', and Hw = 〇. Therefore, Μ1=0, M2=q, Μ3=0, and the voltage is _1-V3. The V phase current command signal is transmitted to the inverter circuit 902 through the multiplex processor, and the inverter circuit 9〇2 is bypassed due to M3=0. The electric vector V3 is transmitted to the space vector modulation 9〇7. The space vector 907 generates a switching control signal S2丨_S26 for selecting to turn any of the driving transistors 601-606 on or off and to transmit an instruction to the pre-driver circuit 5〇4. The pre-driver circuit 504 outputs a signal to the gates of the drive transistors 601-606 in accordance with the output of the SVM control circuit 5〇3 to control the 〇n/〇FF states of the transistors 601-606. Referring to Fig. 25, Fig. 25 is a timing chart showing a pulse width modulation (PWM) switching cycle of an SVM control circuit in accordance with a preferred embodiment of the present invention. The present invention divides a pulse width modulation (PWM) switching period into three states: a detection state, a control state, and a zero state, as shown in FIG. 25: in the detection state, a test voltage vector is Apply for a short period of time to detect critical phase current errors in different areas. Based on the detected phase current error, a suitable voltage vector is selected to control the corresponding phase current. For example, when the detection state is necessary, when the demand output voltage is located in the region I, the voltage vector V3 is transmitted to the space vector modulation module 30 1294717 22134twfl.doc/006 95-11-20 907 to generate switching control. Signal S21_S26, the switching control signals S21-S26 are used to select to turn any of the driving transistors 6〇1-6〇6 on or off. When the SVM control circuit 503 receives the feedback current signal Ifb (ie, Ifb=iv) in the control state soft state, if the V phase current command signal &lt; is greater than or equal to the return, the voltage vector V2 is transmitted to the current signal Ifb. The space vector is modulated 907 to generate switching control signals S21-S26 for selecting to turn any of the driving transistors 601-606 on or off; or, if the v-phase current command signal 4 φ is less than the feedback current signal Ifb, the voltage vector The VI is passed to the spatial vector modulation module 907 to generate switching control signals S2-1-S26 for selecting to turn any of the drive transistors 601-606 on or off. When the v-phase current command 4 is less than the feedback current signal Ifb, once the U-phase current command signal 4 is equal to the feedback current signal Ifb, the SVM control circuit 503 enters a zero state, that is, the voltage vector vo is transmitted to the space vector tone. The change module 907 generates switching control signals S21-S26 for turning off any of the drive transistors 601-606. Similarly, when the V phase current command signal &lt; is greater than or equal to the feedback current signal Ifb, once the W phase current command signal 'equal to the feedback current signal Ifb, the SVM control circuit 503 enters a zero state, that is, the v〇 signal is The transfer to spatial vector modulation module 907 generates switching control signals S21-S26 for turning off any of drive crystals 601-606. Referring to Figure 27, the SVM logic controls the waveform in accordance with a first preferred embodiment of the present invention. The error signal Ie is a comparator 906 that compares the output of the phase current command signal and the feedback current signal Ifb. The feedback current signal Ifb is a DC connection current fed back via a shunt resistor 505. Therefore, according to the above description, the phase current sensed from the DC link current depends on the applied space 1294717 22134twfl.doc/006 95-Π.2 inter-turn vector, which must be multiplexed according to the space vector. And the sign of the feedback current must be determined to account for the nose current error. Figure 28 is a diagram showing the current control waveform of the SVM logic control of the area I in accordance with a first preferred embodiment of the present invention. Figure 29 is a view showing simulation results in accordance with a preferred embodiment of the present invention.

Figure 30 is a diagram showing a variant embodiment of the above preferred embodiment of the present invention. The embodiment of Figure 19 is characterized in that the position detecting circuit adds an output Hk' and the Hk is generated according to the automatic gain control circuit 802U of Fig. 21. The signals of the 802V, 802W outputs H11, H12, and H13 are passed through an x〇R gate, as shown in Figure 31. Figure 32 is a schematic diagram of an SVM control circuit according to a variant embodiment of the present invention. Referring to Figure 24, the main difference is that the lookup table is based not only on the position signals Hu, Hv, Hw, the detection status signal DS, the control status signal CS, and the status. The signal SS also determines the conduction state of the multiplex processor and the output of the spatial vector modulation according to the Hk signal. FIG. 33 shows a pulse width modulation period timing diagram of the SVM control circuit of the embodiment, refer to FIG. In this embodiment, the detection state is divided into 12 states, and the zero state also becomes 2 states. FIG. 34 is a lookup table definition of the SVM control circuit of this embodiment, and FIG. 35 shows the above preferred embodiment of the present invention. As a result of the simulation of the alternative mode, it has been observed that the commutation torque chopping caused by the conventional six-step motor driving technique can be effectively reduced by the technique of the present invention. Furthermore, since only the DC link current is returned by a shunt resistor 505 in the above embodiment, only one hysteresis comparator is required. Based on the detected phase current difference, a suitable space vector is selected for controlling the corresponding phase current with the delay 32 1294717 22134twfl .doc/006 95-11-20. Therefore, by using the predetermined hysteresis band control in the pulse width modulation cycle, an appropriate vector can be selected to use the phase-phase current. Those skilled in the art should understand that the specific embodiments shown in the above drawings and description are merely exemplary and not limiting. The foregoing description of the specific embodiments of the present invention is intended to be exemplary and illustrative. It is not intended to be exhaustive or to enable the present invention in the precise form or the disclosed embodiments. Therefore, the 'previous descriptions should be regarded as exemplary rather than 卞HH f. Many modifications and variations will be apparent to those skilled in the art, and the selection and description of specific embodiments is intended to better explain the present application. The best mode, thus allowing the invention to be practiced in various specific embodiments, and having various modifications suitable for ttr cover implementation. The present invention is intended to be used in the context of the application of the application, and the claims are intended to cover the broadest reasonable scope. It should be understood that changes may be made to the specific embodiments by those skilled in the art, and the scope of the invention as defined in the scope of the claims is not limited. Further, there are no components or components in this disclosure that are intended to be used in the public or whether the components are in the following patent application. The summary of the disclosure is provided by the summary rule of the summary, and the eight final searchers quickly Privately released from this book: reveal the theme. It should be understood that it is not used to explain the scope or meaning of the technology. T-whistle patent 粑 [schematic description] 33 1294717 22134twfl.doc/006 95-11-20 Figure 1 is a line-to-line and line-to-line voltage and phase and line-to-line voltage used with a six-step voltage source inverter Waveform. The second picture shows the voltage waveform of one of the block modulations. Figure 3 shows the phase voltage and current waveform of a six-step voltage source inverter. Figure 4 shows a sinusoidal pulse width modulation (SPWM) technique. Figure 5 is a block diagram of a conventional six stepper motor driver. g Figure 6 is a block diagram of a conventional six step control circuit. Figure 7 shows an analog waveform of a DC link current and a peak 4 sense output current. Figure 8 shows an analog waveform with the current control effect of a conventional six step control architecture. The brother 9 is a block diagram of a modified six step control circuit. Figure 1 shows an analog waveform with the current control performance of the modified six-step control architecture. • The brother map shows the overall control block diagram of a Matsushita method. Figure 12 shows a three-trapezoidal current command for the Matsushita method. The brothers 13(a) to (c) show the current flow of the motor of the first-class Matsushita method.苐14(a) to (b) show the waveform of a Matsushita method. Figure 15 shows a simulation result with the current control performance of the Matsushita method. Figure 16 shows the basic problem of a Matsushita method with a trapezoidal current waveform. ι·- S y 34 1294717 22134twfl.doc/006 95-11-20 Figure 17 shows the basic problem of the Matsushita method with sinusoidal current waveforms. Figure 18 shows the experimental results of the Matsushita method under (a) low speed operation and (b) high speed operation. Figure 19 is a block diagram of a motor driver in accordance with a preferred embodiment of the present invention. Figure 20(a) shows the definition of the space vector and the switch state mode. φ 20(b) is a predetermined waveform for one of the phase currents in accordance with a preferred embodiment of the present invention. Figure 21 is a diagram showing a position detecting circuit and a current command generating circuit in accordance with a preferred embodiment of the present invention. Figure 22 shows a phase offset table. Figure 23 is a diagram showing the output waveform of the position detecting and current command generating circuit in accordance with a preferred embodiment of the present invention. Figure 24 shows a schematic diagram of an SVM control φ circuit in accordance with a preferred embodiment of the present invention. Figure 25 is a timing chart showing an SVM control circuit in accordance with a first preferred embodiment of the present invention. Figure 26 shows a look-up table of an SVM control circuit in accordance with a first preferred embodiment of the present invention. Figure 27 is a diagram showing an SVM logic control waveform in accordance with a first preferred embodiment of the present invention. Figure 28 is a diagram showing the current control waveform of the SVM control circuit in the area according to a first preferred embodiment of the present invention. 35 1294717 22134twfl.doc/006 95-11-20 Figure 29 shows a simulation result in accordance with a preferred embodiment of the present invention. Figures 30-35 show a variation of the above preferred embodiment of the invention. [Main component symbol description] 101, 103, 105, 601, 603, 605: upper side driving transistors 102, 104, 106, 602, 604, 606: lower side driving transistors 101D, 102D, 103D, 104D, 105D, 106D , 3D, 5D, Lu 601D, 602D, 603D, 604D, 605D, 606D: diode 201, 501: Hall sensor circuit 202: six-step square wave control circuit 203: pre-drive circuit 204: current detection Resistor 300 '700 : motor 301, 302, 303: phase coil IA, IB, 1C, ia, ib, ic: phase current art 201A, 201B., 201C, 501A, 501B, 501C: Hall sensor H1+, HI-, H2+, Η2·, H3+ and H3-: Hall sensor wheeled

Tc: torque command signal

Ifb: feedback current signals S11-S16, S21-S26: switching control signals 401A, 401B, 401C, 801U, 801V, 801W: differential amplifiers 402A, 402B, 402C, 802U, 802V, 802W: automatically select only 36 1294717 22134twfl. Doc/006 95-11-20 Control circuits 403A, 403B and 403C, 408, 806U, 806V, 806W · Adders 404A, 404B and 404C, 805a-805f, 807U, 807V, 807W: Multipliers 405A, 405B, 405C 412A, 412B, 412C, 804U, 804V, 804W, 906: Comparator 0 406A, 904: Low pass filter 407: Peak detection circuit 409: Controller 410: Carrier signal generator 411: Invalid time control circuit FI, F2: control signal TP: ramp current command TU1: time interval g 502: position detecting circuit and current command generating circuit 503: space vector modulation control circuit 504: pre-drive circuit 505: current detecting resistor 701: U phase Coil 702: V-phase coil 703: W-phase coil Vcc: Power supply Iu: U-phase current CS &gt; 37 1294717 95-11-20 22134twfl.doc/006 Ιγ · v phase current Iw · Phase current Hu, Hv, Hw: Position signal /\ : U phase current command No.: V phase current command signal: W phase current command signal V0~V7: voltage vector 803U, 803V, 803W: level offset circuit ® H11, H12, H13: signal 808: phase offset table 809: torque amplitude ratio Gain control circuit I, K2: value 901: multiplex processor 902: inverter circuit 903: level shift circuit _905: level shift and amplifier 907: space vector modulation 908: reference clock generator 909, 911: D-type flip-flop 910: delay 912: falling edge delay 913, 916: inverter 914: NAND gate 917: look-up table 38 1294717 95-11-20 22134twfl.doc/006 DS: detection status signal CS: Control status signal SS: status signal S21-S26: switching control signal I~VI: area

39

Claims (1)

96-11-15 1294717 22134twf2.doc/006 X. Patent application scope: 1. A motor driver for driving a motor, the motor comprising a plurality of stator coils, one end of each stator coil being commonly coupled to a common node The motor driver comprises: a plurality of output circuits, each output circuit comprising an upper arm switch and a lower arm switch, wherein a connection point between the upper arm switch and the lower arm switch of the i-th output circuit is coupled to the i-th stator coil a current detecting resistor coupled to the plurality of output circuits; a position detecting circuit for outputting a position signal related to a position of the motor rotor; and a current command generating circuit for determining the position signal and the The predetermined phase angle generates a plurality of predetermined current command signals to correspond to the stator coils; and a space vector modulation control circuit receives the predetermined current command signals and couples the current detecting resistors when the kth predetermined current command Controlling the upper arm* switch and the lower arm switch of the output circuits to receive the power when the phase of the signal is positive or negative and a predetermined phase The current of the kth stator coil drawn by the current detecting resistor compares the current of the kth stator coil with the magnitude of the kth predetermined current command signal to select one of the stator coils and control the current magnitude thereof, wherein i and k are natural numbers. 2. The motor driver according to claim 1, wherein the position detecting circuit has a three-position detecting circuit, each of which comprises: a differential amplifier that receives a plurality of Hall sensors 1294717 22134twf2 .doc/006 year-and-month repair (more) is replacing page 96·11, 1 5 _________________ 96-11-15 output for obtaining a differential output of the complex outputs of the Hall sensors; a quasi-offset circuit for shifting a voltage level of the differential output; and a comparator for outputting a position signal. 3. The motor driver of claim 2, wherein the position detecting circuit further comprises at least one automatic gain control circuit for adjusting the output of the differential amplifier to have the same peak value. 4. The motor driver of claim 2, wherein the current command generating circuit comprises a phase shift table for receiving the predetermined phase angle for determining a first gain value and a second gain a first current command generating circuit, comprising: a first multiplier receiving a first differential output of the Hall sensor output and the first gain value for generating a first signal; And a second multiplier receiving a first differential output of the Hall sensor output and the second gain value for generating a second signal; a second current command generating circuit, including a third multiplier receiving a second differential output of the Hall sensor output and the first gain value for generating a third signal; and a fourth multiplier receiving the Hall (Hall) sensor output 41 1294717 22134twS.doc/006 month repair (more) positive replacement page QfV H LS_--- 96-11-15 a second differential output and the second gain value for Generating a fourth signal; generating a third current command The circuit includes: a fifth multiplier receiving a third differential output of the Hall sensor output and the first gain value for generating a fifth signal; a sixth multiplier receiving a third differential output of the Hall sensor output and the second gain value for generating a sixth signal; a first adder receiving the fifth signal and the second signal for Generating a first phase angle signal; a second adder receiving the first signal and the fourth signal for generating a second phase angle signal; a third adder receiving the third signal and the sixth a signal for generating a third phase angle signal; a seventh multiplier receiving a torque command signal and the first phase angle signal for generating a first predetermined current command signal; an eighth multiplier receiving The torque command signal and the second phase angle signal are used to generate a second predetermined current command signal; and a ninth multiplier receives the torque command signal and the third phase angle signal for generating a first Three predetermined current command letter number. 5. The motor driver of claim 4, wherein the current command generating circuit further comprises a torque amplitude proportional gain control circuit for adjusting the value of the torque command signal. 42 1294717 22134twf2.doc/006 Year of the month: A more) is replacing the page "1" - 96-11-15 6. A motor controller that controls multiple output circuits, each of which contains a An upper arm switch and a lower arm switch, wherein a connection point between the upper arm switch and the lower arm switch of the i-th output circuit is coupled to the i-th stator coil, the controller comprising: a current detecting resistor coupled to the above a plurality of output circuits; a position detecting circuit for outputting a position signal related to the position of the motor rotor; a current command generating circuit for generating a plurality of predetermined current command signals according to the position signal and a predetermined phase angle, And corresponding to the stator coils; and a space vector modulation control circuit, receiving the predetermined current command signals and coupling the current detecting resistors, when the phase of the kth predetermined current command signal is at a positive or negative predetermined phase Controlling the upper arm switch and the lower arm switch of the output circuits to receive the current of the kth stator coil captured by the current detecting resistor, comparing the current of the kth stator coil with the kth pre Magnitude of the current command signal to select one from the plurality of stator coils, and to control the magnitude of the current, wherein i and k is a natural number. 7. The controller of claim 6, wherein the position detecting circuit has a three-position detecting circuit, each of which includes: a differential amplifier that receives a complex output of a Hall sensor. Obtaining a differential output of the complex outputs of the Hall sensors; a quasi-offset circuit for shifting a voltage level of the differential output 43 1294717 22134twf2.doc/006 The replacement page ^11·15 an 1 5 is normal; and a comparator for outputting a position signal. 8. The controller of claim 7, wherein the position detecting circuit further comprises at least one automatic gain control circuit to adjust the outputs of the differential amplifiers to have the same peak value. 9. The controller of claim 6, wherein the current command generating circuit comprises: a phase shifting device that receives the predetermined phase angle for determining a first gain value and a second gain value; a first current command generating circuit, comprising: a first multiplier receiving a first differential output of the Hall sensor output and the first gain value for generating a first signal; and a a second multiplier receiving a first differential output and a second gain value of the Hall sensor output for generating a second signal; a second current command generating circuit comprising: a first a three multiplier, receiving a second differential output of the Hall sensor output and the first gain value for generating a third signal; and a fourth multiplier receiving the Hall a second differential output of the sensor output and the second gain value for generating a fourth signal; a third current command generating circuit comprising: 44 1294717 22134twG.doc/006 a fifth multiplier (more) positive replacement page 96 · η_15 a Hall receives the third differential output (Hall) of the sensor output of the first gain value and for generating a fifth channel a sixth multiplier, receiving a third differential output of the Hall sensor output and the second gain value for generating a sixth signal; a first adder receiving the fifth signal and The second signal is used to generate a first phase angle signal; a second adder receives the first signal and the fourth signal for generating a second phase angle signal; and a third adder receives the signal The third signal and the sixth signal are used to generate a third phase angle signal; a seventh multiplier receives a torque command signal and the first phase angle signal for generating a first predetermined current command signal; An eighth multiplier receiving the torque command signal and the second phase angle signal for generating a second predetermined current command signal; and a ninth multiplier receiving the torque command signal and the third phase angle a signal for generating a third predetermined current command signal. 10. The controller of claim 9, wherein the current command generating circuit further comprises a torque amplitude proportional gain control circuit for adjusting the torque command signal value. 11. A method for controlling an electric motor, the electric motor having a plurality of stator coils coupled to a plurality of output circuits, the output circuit couplings 45 96-11-15 1294717 22134twf2.doc/006 In connection with a current detecting resistor, the method comprises the steps of: providing a plurality of predetermined current command signals respectively corresponding to the stator coils; and when the kth predetermined current command signal The phase is positive or negative for a predetermined period of time, wherein the specific period includes at least one detection state period and a control state period: during the detection state, the output circuits are controlled to cause current to flow through the kth stator coil And detecting a current of the kth stator coil through the current detecting resistor when entering the control state period from the detecting state, and comparing the current to the kth predetermined current command signal to The stator coils select a particular stator coil and control the magnitude of the current of the particular stator coil, where k is a natural number. 12. The method for controlling an electric motor according to claim 11, further comprising: controlling the output circuits to stop when a current flowing through the specific stator coil is equal to a predetermined current command signal corresponding thereto The stator coils are supplied with electric power. 13. The method for controlling an electric motor according to claim 11, wherein the stator coils comprise a first phase, a second phase and a third phase stator coil, and the predetermined current command signals include A first phase, a second phase and a third phase predetermined current command signal. 14. The method for controlling an electric motor according to claim 13, wherein the first phase, the second phase, and the third phase predetermined current command letter are replaced by a (more) replacement page. 1294717 22134twf2.doc/006 is a sine wave with a phase difference of 120 degrees. 15. The method for controlling an electric motor according to claim 13, wherein the preset phase is 30 degrees, and the specific period is the k-th phase predetermined current command signal from a negative 30 degree phase to a positive 30 degree A period formed by the phase or a period in which the k-th phase predetermined current command signal is composed of a positive 30-degree phase to a negative 30-degree phase. 47