Method for detecting excitation position of motor

Abstract

The invention provides a motor excitation position detection method capable of reliably detecting the excitation position of a motor and operating at a low speed even if the driving voltage varies when a three-phase brushless motor is operated at a low speed in a sensorless driving manner by performing PWM control through 120-degree energization. When the drive voltage varies, the motor inductance is Lm, the resistance value is Rm, the drive voltage is V, the conduction time is t, and the motor current is Im, and the following equation is satisfied: [mathematical equation 1] # imgabs0 # MPU (51) substitutes the conduction time t1 obtained from a predetermined drive voltage V1, a PWM period, and a predetermined duty ratio into equation 1 to obtain the motor current Im; [mathematical formula 2] t = (Lm / Rm) log (-V / ImRm-V)... (formula 2): the motor current Im obtained by (formula 1) and the driving voltage V2 obtained by changing the driving voltage V1 are substituted into (formula 2), the conduction time t2 is calculated, the conduction time t2 is divided by the PWM period, the duty ratio is calculated and updated, and the inverter circuit (52) applies the driving voltage V2 at the updated duty ratio to perform operation.

Description

Technical Field

[0001] The present invention relates to a method for detecting the excitation position of a motor when the motor rotates at a low speed. Background Art

[0002] Conventionally, small DC motors used brushed DC motors, but due to problems such as brush noise, electrical noise, and durability, brushless DC motors have emerged. In addition, recently, from the viewpoints of miniaturization, light weight, firmness, and low cost, sensorless motors without position sensors have attracted attention and have first been adopted in hard disk drives and the like in the information equipment field. With the development of vector control technology, they have also begun to be adopted in the household electric appliance and vehicle-mounted fields.

[0003] In the sensorless drive method, the rotor position is detected based on the induced voltage. However, since no induced voltage is generated when stationary, the rotor position cannot be grasped, and thus the motor cannot be started. In order to detect the rotor position when stationary, there is a method of providing a coil current sensor and a current detection circuit, flowing a sinusoidal coil current through the coil by PWM driving using an inverter, and estimating the position based on the current response. In addition, the rotor position at rest can be detected based on the inductance deviation using the above-described typical method. Alternatively, the position can be determined by forcibly rotating the rotor by forced commutation without performing position sensing.

[0004] However, once starting, energization for rotation is performed, so the method of applying a sensing pulse and detecting the rotor position based on the inductance deviation becomes difficult. For example, it can be considered to superimpose a high-frequency current on the excitation current to detect the inductance deviation, but both the scale of the hardware and software will increase. In addition, the influence of magnetic saturation or induced voltage must be considered, and there are also difficult-to-estimate factors such as the inherent errors of the motor and the drive circuit. Therefore, after forcibly positioning the rotor by fixed excitation without performing position detection, a ramp start method of gradually increasing the rotational speed while maintaining synchronization is widely adopted. However, this method has problems that the positioning of the rotor takes a long time and further reverse rotation is required. In addition, since synchronization is achieved by open-loop control, there are also disadvantages that acceleration takes time and it is easy to get out of synchronization due to load changes. To avoid this situation, starting with a large current results in reduced efficiency and a larger DC power supply. Since a misalignment occurs during load changes, the application is limited and it cannot be used in reciprocating motion mechanisms, applications where rotation is caused by an external force, viscous loads, or applications with load changes.

[0005] Therefore, as a method for detecting the excitation position that uses simple hardware and software to achieve cost reduction, does not generate a sensing sound during startup, and can detect the rotor position in units of 120° energization intervals, the following method has been proposed: Utilize the fact that in the 120° energization drive of a three-phase brushless motor, if a voltage is applied to the motor, the induced voltage generated in the non-energized phase changes according to the position of the permanent magnet excitation (rotor) to detect the position of the rotor (Patent Document 1: Japanese Patent Laid-Open No. 2019-17235). Prior Art Documents Patent Documents

[0006] Patent Document 1: Japanese Patent Laid-Open No. 2019-17235 Summary of the Invention Technical Problems to be Solved by the Invention

[0007] However, when the motor is energized in a 120° energization manner, the induced voltage generated in the non-energized phase changes according to the magnitude of the drive voltage applied to the motor. Especially in the case of PWM drive, it changes according to the magnitude (energization time) of the duty ratio of the drive voltage. Figure 10 It is a graph showing the change in the induced voltage generated in the non-energized phase according to the rotor rotation position when the drive voltage varies at a duty ratio of 20% (fixed) for a certain motor. According to this graph, if there is a voltage variation of a certain level or more due to a change in the battery system or power supply voltage, etc., the waveform of the induced voltage may change significantly and it may be impossible to detect the excitation position.

[0008] In Patent Document 1, in order to eliminate this phenomenon, a method of multiplying the duty ratio by a correction coefficient was proposed. However, depending on the motor, even if the induced voltage is multiplied by the correction coefficient, it is difficult to detect the position at a low duty ratio below a certain level or a high duty ratio above a certain level. Therefore, only a limited range of duty ratios can be used. Technical Means for Solving the Technical Problems

[0009] The present invention is proposed to solve these problems, and its object is to provide a method for detecting the excitation position of a motor, which can reliably detect the excitation position of the motor and operate at a low speed even when the drive voltage varies when the three-phase brushless motor is driven in a sensorless manner at a low speed by PWM control through 120° energization.

[0010] A method for detecting the excitation position of a motor, the motor used includes: a rotor with permanent magnet excitation; a stator with three-phase coils; an output unit that supplies power to the three-phase coils bidirectionally via a half-bridge inverter circuit; a control unit that performs PWM control on the coil output according to instructions from an upper controller, stores energization angle information and energization direction information in units of 60° energization intervals that can rotate continuously, and performs switching control on the output unit based on this information to switch the energization state; and a measurement unit that performs A / D conversion on the three-phase coil voltage and sends it to the control unit. The control unit periodically performs 120° energization including a cut-off cycle in the energization direction where the position where self-excitation stops by two-phase fixed energization via the output unit coincides with the starting position of the 60° energization interval. The measurement unit measures the energized phase voltage and the non-energized phase voltage during the conduction cycle of PWM energization, so as to detect the excitation position of the motor while operating in a sensorless manner. When the drive voltage changes during operation with a drive voltage applied to the three-phase coils at a specified duty ratio, if the inductance of the motor is Lm, the resistance value is Rm, the drive voltage is V, the energization time is t, and the motor current is Im, the following formula holds: [Mathematical formula 1] The control unit substitutes the energization time t1 obtained according to the specified drive voltage V1, the period of PWM, and the specified duty ratio into (Formula 1), calculates the motor current Im, and solves (Formula 1) for t to obtain: [Mathematical formula 2] t = (Lm / Rm)log e (-V / ImRm - V)... (Formula 2) Substitute the motor current Im obtained from (Formula 1) and the drive voltage V2 after the change of the drive voltage V1 into (Formula 2), calculate the energization time t2, divide the energization time t2 by the PWM period, calculate and update the duty ratio, and the control unit operates by applying the drive voltage V2 with the updated duty ratio through the output unit.

[0011] In this way, when the drive voltage changes to V2 during low-speed operation with a drive voltage V1 applied to the three-phase coils at a specified duty ratio, the following formula is used: [Mathematical formula 1] The control unit substitutes the energization time t1 obtained according to the specified drive voltage V1, the period of PWM, and the specified duty ratio into (Formula 1), calculates the motor current Im, and solves (Formula 1) for t to obtain: [Mathematical formula 2] t = (Lm / Rm)log e (-V / ImRm - V)...(Equation 2) Substitute the motor current Im obtained from (Equation 1) and the drive voltage V2 after the drive voltage V1 changes into (Equation 2), calculate the energization time t2, divide this energization time t2 by the PWM period, calculate and update the duty ratio, and the control unit applies the drive voltage V2 with the updated duty ratio through the output unit. The measurement unit measures the energized phase voltage and the non-energized phase voltage during the conduction cycle of the PWM energization to detect the excitation position of the motor. Thereby, it is possible to detect the excitation position of the motor and perform low-speed operation without being affected by the change in the drive voltage. Advantages of the Invention

[0012] The present invention can provide a method for detecting the excitation position of a motor. When performing PWM control in a 120° energization mode to make a three-phase brushless motor operate at low speed, even if the drive voltage changes, it is possible to reliably detect the excitation position of the motor and perform low-speed operation by updating the duty ratio. Description of the Drawings

[0013] Figure 1 It is a waveform diagram of the inductance and the non-energized phase coil voltage during U-V excitation. Figure 2 It is a waveform diagram of the inductance and the non-energized phase coil voltage during U-W excitation. Figure 3 It is a waveform diagram of the inductance and the non-energized phase coil voltage during V-W excitation. Figure 4 It is a waveform diagram of the inductance and the non-energized phase coil voltage during V-U excitation. Figure 5 It is a waveform diagram of the inductance and the non-energized phase coil voltage during W-U excitation. Figure 6 It is a waveform diagram of the inductance and the non-energized phase coil voltage during W-V excitation. Figure 7 is a measured waveform diagram of the non-energized phase coil voltage. Figure 8 It is a structural block diagram of the drive circuit of a three-phase brushless DC motor. Figure 9 It is a structural diagram of a star-connected three-phase brushless DC motor. Figure 10 It is a waveform diagram of the induced voltage generated by the non-energized phase coil when the drive voltage changes with the duty ratio of the PWM drive fixed at 20%. Figure 11 It is a waveform diagram of the induced voltage generated by the non-energized phase coil when the duty ratio is corrected according to the voltage change from the drive voltage with a duty ratio of 20% in the PWM drive. Detailed Embodiment

[0014] Hereinafter, an embodiment of a method for detecting the excitation position of a motor according to the present invention will be described with reference to the accompanying drawings. As an example of the motor, a sensorless motor having a permanent magnet excitation on the rotor, windings arranged with a 120° phase difference on the stator and star-connected, and the phase terminals connected to the motor output unit will be used for the description.

[0015] Hereinafter, as an example, a method for detecting the permanent magnet excitation position of a sensorless motor for sensorless driving of a three-phase DC brushless motor will be described together with the structure of a sensorless motor driving device. Refer to Figure 9 FIG. shows an embodiment of a three-phase brushless DC motor according to the present invention. As an example, a three-phase brushless DC motor having a 2-pole permanent magnet rotor and a stator 4 provided with 3 slots is illustrated. The motor can be an inner rotor type or an outer rotor type. In addition, as the permanent magnet type excitation, it can be any one of an interior permanent magnet type (IPM type) motor and a surface permanent magnet type (SPM type) motor.

[0016] In Figure 9 a rotor 2 is integrally provided on a rotor shaft 1, and a 2-pole permanent magnet 3 is provided as excitation. On the stator 4, pole teeth U, V, and W are arranged opposite to the permanent magnet 3 with a 120° phase difference. Windings u, v, and w are provided on the respective pole teeth U, V, and W of the stator 4, and are star-connected at a common point C between phases, forming a three-phase brushless DC motor wired to a motor driving device described later. In addition, since a common line is not required, it is omitted.

[0017] Next, Figure 8 FIG. shows an example of a drive circuit for a three-phase DC brushless motor. As a drive method at startup, 120° energization bipolar rectangular wave excitation is assumed. MOTOR is a three-phase sensorless motor. MPU51 is a microcontroller (control unit). MPU51 stores excitation position information and performs switching control of the output unit according to a rotation command RUN from an upper controller 50, arbitrarily switching the excitation state, where the excitation position information specifies six energization directions for the three-phase coils (U, V, W) and excitation switching intervals (interval 1 to interval 6) of 120° energization corresponding to each energization direction.

[0018] The inverter circuit 52 (INV: output unit) energizes the three-phase coils and performs switching operations such as excitation phase switching or PWM control to control the motor torque. The inverter circuit 52 includes diodes reversely connected in parallel with the switching elements, and a half-bridge type switching circuit capable of arbitrarily connecting to the positive power supply line and the ground power supply line is provided for each of the three phases. The current sensor 53 measures the coil current when the three-phase coil is PWM-powered. Specifically, a shunt resistor r is provided between the common ground side terminal of the inverter circuit 52 and the ground. Since only a low voltage of several V equivalent to the voltage drop is applied to the shunt resistor r, it can be used even when the coil applied voltage is a high voltage of several hundred V. The operational amplifier 54 amplifies the coil voltage corresponding to the coil current and sends it to the A / D conversion circuit 55 (ADC: measurement unit).

[0019] The A / D conversion circuit 55 is connected to the coil output terminals U, V, and W, samples the coil voltages of each phase simultaneously using the conversion start signal from the MPU 51, performs analog-to-digital conversion in sequence, and sends the conversion result to the MPU 51. Usually, the ADC 55 is built into the MPU 51. In the case of using the built-in ADC 55, since the maximum input voltage is low, it is preferable to provide a resistive voltage division circuit. In this way, the drive circuit can be configured very simply.

[0020] The inductance change (spatial high-order harmonics) caused by the rotor angle θ is known to be approximately ΔL = -cos(2θ), and each electrical angle has two periods. On the other hand, it is known that when the three-phase coil is two-phase powered by rectangular wave PWM power supply, two-periodic voltage fluctuations are observed in the non-powered phase centered on the neutral point potential according to θ.

[0021] Figure 1 It shows the theoretical value waveforms of the voltage change waveform ΔVw of the non-powered phase, the inductance changes (ΔLu, ΔLv) of the U-phase and V-phase, and the combined inductance change ΔLu-v of the two phases when rotating one electrical angle while performing U-V excitation by PWM power supply. The voltage change waveform is obtained by inverting the polarity of the combined inductance change waveform, and performs positive and negative oscillations centered on the neutral point potential of half of the coil applied voltage.

[0022] Figure 7A It shows the measured waveform of the non-powered phase coil voltage of an inner-rotor type motor. The theoretical value of the voltage waveform of the non-powered phase reflects the inductance and becomes the opposite polarity, but the waveforms are roughly approximated, indicating that the assumption is correct. In addition, in the case of rectangular wave power supply, ringing occurs in the induced voltage. However, as a result of the measurement, the ringing time is very short, and it converges within the range of measurement error in several μs to several tens of μs in various motors. Even for the rectangular wave PWM power supply pulse for motor drive, the induced voltage can be detected with high precision.

[0023] If a large current flows through the three-phase coil, magnetic saturation will occur, and the inductance will not change. This is particularly significant in small outer-rotor motors. When magnetic saturation occurs, in the two periodic inductance change waveforms, the peaks and valleys adjacent to the excitation (Setup) position where self-excitation stops due to two-phase fixed power supply remain, but the other peaks and valleys disappear, becoming one periodicity. Figure 7B Fig. shows an example of the inductance waveform that becomes one periodicity due to magnetic saturation. The motor used in the measurement is a small outer-rotor motor, different from the motor used in Figure 7A . The excitation position during U-V power supply is 150°. In the ΔVw waveform, only the peaks and valleys adjacent to the excitation position are clearly observed.

[0024] The excitation position where self-excitation stops due to two-phase fixed power supply is the zero-crossing point of the inductance and also the zero-crossing point of the induced voltage. The excitation point and the adjacent peaks and valleys are also stable for magnetic saturation. As can be seen from Figure 7A and Figure 7B , the voltage variation of the non-powered phase reflects the rotor angle θ, and monotonicity is guaranteed within the interval. Therefore, even when no induced voltage is generated at rest, the excitation current can flow to estimate the rotor position. The voltage variation amplitude is more than 10% of the voltage applied to the coil and is also in the order of several volts. Considering that the induced voltage in the millivolt level is detected during startup in the existing method, it has an overwhelming advantage.

[0025] As described above, by using rectangular wave PWM control for motor drive to detect the inductance change of the non-powered phase coil and only using the inductance change near the excitation position, stable position detection can be performed from the stationary state to the low-speed rotation region. Thereby, the sensing steps are simplified, no power is required during sensing, the efficiency is improved, and no sensing noise is generated to achieve quiet operation.

[0026] The angles and power supply directions of each power supply interval for 120° power supply are summarized in the following table. The CW power supply in the table is the power supply direction for rotation in the direction of increasing angle, and the CCW power supply is the power supply direction for rotation in the direction of decreasing angle. The excitation power supply is the power supply direction that stops self-excitation at the angle described in the () within the frame in the table. Both the start point and the end point are recorded for each interval. In each power supply direction, the phase connected to the + power supply side is recorded first, and the phase connected to the GND side is recorded after the hyphen.

[0027]

Table 1

[0028] Let the rotation direction in which the interval number described in Table 1 increases be CW, and the rotation direction in which it decreases be CCW. When rotating in the CW direction, the end position of the interval is the adjacent + side interval, and when rotating in the CCW direction, the end position of the interval is the boundary point with the - side interval. For example, in the case of interval 1, when rotating in the CW direction, the end position of the interval is the boundary point with interval 2 at 90°, and when rotating in the CCW direction, the end position of the interval is the boundary point with interval 6 at 30°.

[0029] In Figure 1 in the case of U-V excitation, when rotating in the CW direction, the starting point of the energized interval is point A, and the ending point of the energized interval is point B. The excitation starting point is point C, and the phase of the trough part where point B is located is stable and can be used for position detection. Therefore, positive and negative threshold values Vth with a specified potential difference relative to the neutral point potential are preset, and the non-energized phase coil voltage ΔV w is compared with the threshold value Vth in each measurement. If it exceeds the threshold value, it can be detected that the interval end point has been exceeded. When rotating in the CCW direction, V-U power supply is applied, so refer to Figure 4 . The rotor rotates from the electrical angle 90° side to the electrical angle 30° side. Therefore, the interval end point is the electrical angle 30°. Since the excitation starting point is the electrical angle 330°, the trough part on the electrical angle 30° side has a stable phase and can be used for position detection. Therefore, similar to the CW case, by comparing the voltage of the non-energized phase W with the threshold value Vth, the interval end point can be detected.

[0030] In interval 2 where the electrical angle is from 90° to 150°, U-W excitation is selected. Figure 2 represents the inductance change and the change of the non-energized phase coil voltage during U-W excitation. The waveform is obtained by shifting the waveform of Figure 1 by 60° and reversing the polarity. The non-energized phase is the V phase, and the excitation position C point is the electrical angle 210°. When rotating in the CW direction in interval 2, the non-energized phase coil voltage will necessarily pass through point B. Therefore, the rotor position at this moment is the electrical angle 150°. If it is detected that after point B, it is switched to interval 3, continuous rotation can be achieved.

[0031] In interval 3 where the electrical angle is from 150° to 210°, V-W excitation is selected. Figure 3 represents the inductance change and the change of the non-energized phase coil voltage during V-W excitation. The waveform is obtained by shifting the waveform of Figure 2 by 60° and reversing the polarity. The non-energized phase is the U phase, and the excitation position C point is the electrical angle 270°. When in the case of being located in interval 3 and rotating in the CW direction, the non-energized phase coil voltage necessarily passes through point B. Therefore, the rotor position at this moment is an electrical angle of 210°. If it switches to interval 4 after detecting point B, continuous rotation can be achieved.

[0032] In interval 4 where the electrical angle ranges from 210° to 270°, V-U excitation is selected. Figure 4 Represents the inductance change and the non-energized phase coil voltage change during V-U excitation. The waveform is obtained by shifting the waveform of Figure 3 by 60° and reversing the polarity. The non-energized phase is phase W, and the excitation start position C is at an electrical angle of 330°. When in the case of being located in interval 4 and rotating in the CW direction, the non-energized phase coil voltage necessarily passes through point B. Therefore, the rotor position at this moment is an electrical angle of 270°. If it switches to interval 5 after detecting point B, continuous rotation can be achieved.

[0033] In interval 5 where the electrical angle ranges from 270° to 330°, W-U excitation is selected. Figure 5 Represents the inductance change and the non-energized phase coil voltage change during W-U excitation. The waveform is obtained by shifting the waveform of Figure 4 by 60° and reversing the polarity. The non-energized phase is phase V, and the excitation start position C is at an electrical angle of 30°. When in the case of being located in interval 5 and rotating in the CW direction, the non-energized phase coil voltage necessarily passes through point B. Therefore, the rotor position at this moment is an electrical angle of 330°. If it switches to interval 6 after detecting point B, continuous rotation can be achieved.

[0034] In interval 6 where the electrical angle ranges from 330° to 30°, W-V excitation is selected. Figure 6 Represents the inductance change and the non-energized phase coil voltage change during W-V excitation. The waveform is obtained by shifting the waveform of Figure 5 by 60° and reversing the polarity. The non-energized phase is phase U, and the excitation start position C is at an electrical angle of 90°. When in the case of being located in interval 6 and rotating in the CW direction, the non-energized phase coil voltage necessarily passes through point B. Therefore, the rotor position at this moment is an electrical angle of 30°. If it switches to interval 1 after detecting point B, continuous rotation can be achieved.

[0035] In this way, the interval end points located at the peak or trough adjacent to the excitation start position can be detected by a preset threshold. Then, if the interval number after the non-energized phase coil voltage exceeds the threshold is +1 during CW rotation and -1 during CCW rotation, continuous rotation can be achieved.

[0036] Similar to the above, assume that the rotation direction in which the interval number increases is CW, and the rotation direction in which it decreases is CCW. The starting position of the interval in the CW case is the adjacent -(negative) side interval, and the starting position of the interval in the CCW case is the boundary point with the +(positive) side interval. For example, in the case of the energized interval 1, the starting position of the interval in the CW case is the boundary point 30° with the energized interval 6, and the starting position of the interval in the CCW case is the boundary point 90° between the energized interval and interval 2.

[0037] Figure 1 In, when rotating in the CW direction in the case of U-V energization, the starting point of the interval is represented by point A. During normal operation, since it rotates in the desired rotation direction, detection of the starting point of the interval is not required. However, when rotating at a low speed in the direction opposite to the desired rotation direction using an external force, starting point detection is required for accurate field switching. Braking deceleration is required during high-speed rotation, and the detection of the starting point position only considers low-speed rotation. In the case of rotating in the opposite direction, the induced voltage becomes a problem. In Figure 1 the induced voltage of the non-energized phase W is at the zero-crossing point at the electrical angle 60° at the center of the interval. When rotating in the forward rotation direction, the gradient caused by the inductance change is consistent with the gradient of the induced voltage, and the end point of the interval can be reliably detected. However, when rotating in the reverse direction, the gradients of the two are opposite, and the waveform caused by the inductance change is cancelled out, making it difficult to detect the starting point of the interval. In addition, in the case of a motor that becomes one cycle due to magnetic saturation, it is almost impossible to detect the starting point of the interval.

[0038] Therefore, if we focus on Figure 5 the excitation starting position C point of the W-U excitation, since point C passes through the electrical angle 30°, by comparing the magnitudes of the neutral point potential and the non-energized phase V phase voltage ΔVv, if ΔVv is less than the neutral point potential, it is possible to detect the situation of rotating more than 30° electrical angle towards the interval 6 side. Therefore, when performing U-V excitation in interval 1, if it is switched to W-U excitation for an instant and the non-energized phase V phase voltage is measured, it is possible to determine whether it is before or after the electrical angle 30°. If it is periodically measured repeatedly before passing through the electrical angle 30°, the starting point of the interval, that is, the field switching position, can be detected.

[0039] For the detection of the starting point electrical angle 90° when rotating in the CCW direction, similar to the CW direction rotation, referring to Figure 6 perform W-V excitation, and by measuring the non-energized phase U phase voltage, the electrical angle 90° position can be detected. Before and after the excitation starting position, that is, the electrical angle 30° or the electrical angle 90°, the voltage change gradient is steep, making it easy to determine positive and negative, and the phase shift is also small. Therefore, position detection can be reliably performed. Although the sensing consumes power very little, it is desirable to extend the sensing period.

[0040] The same applies to the energized intervals 2 to 6. If the energization direction that becomes the excitation start point is selected to periodically detect the inductor zero crossing, the start point of the interval can be detected. When the start point is detected, since it is rotating in the reverse direction, continuous rotation can be achieved as long as the interval number is reversed backward.

[0041] In the case where the rotor reverses at an extremely low speed in the direction opposite to the desired rotation direction due to an external force, etc., in order to resume forward rotation, it is necessary to detect the start point of the interval and perform excitation switching, and the start point of the interval can be detected by setting a start threshold. For example, in Figure 1 , when it is in interval 1, it is sufficient to set the potential at point A as the start threshold. Figures 2 - 6 The same applies to intervals 2 to 6 of , and it is sufficient to set the potential at point A of each interval as the start threshold. Therefore, a start threshold Vth2 having a prescribed potential difference with respect to the neutral point potential is preset, and the non-energized phase coil voltage ΔV and the start threshold Vth2 are compared in magnitude each time a measurement is made. If the start threshold is exceeded, the start of the interval can be detected. Additionally, the gradient of the non-energized phase coil voltage ΔV can be determined. If the gradient is opposite to that during forward rotation, it can be detected as a reverse rotation state.

[0042] Therefore, if the start point is detected while the rotor is in the reverse rotation state, the excitation interval is retreated by one interval for excitation, thereby generating a forward torque, that is, braking to suppress reverse rotation, and it is possible to resume forward rotation. However, since the polarity of the induced voltage during reverse rotation is opposite to that during forward rotation, the non-energized phase coil voltage ΔV at the start point of the interval is smaller than that during forward rotation and does not exceed the start threshold Vth2. In this case, the induced voltage can be estimated by calculation and the start threshold Vth2 can be corrected. Or, the start point detection can be limited to extremely low speed rotation where the error caused by the induced voltage can be ignored.

[0043] According to this method, without performing special excitation position detection excitation, the start point can be detected in the drive excitation state. Thereby, the energization efficiency is not reduced, and the electromagnetic noise caused by sensing energization is not generated. Additionally, by detecting the start point, braking can be performed from the reverse rotation state to resume forward rotation.

[0044] The change in the non-energized phase coil voltage varies according to the duty ratio of the drive voltage. When the motor is driven with a prescribed drive voltage and a constant duty ratio, if a voltage variation of a certain level or more occurs due to a change in the battery system or power supply voltage, etc., as Figure 10As shown, the waveform of the induced voltage induced in the non-energized phase coil may change significantly and it may be impossible to detect the excitation position. Therefore, as described below, a method is adopted in which the duty ratio of the drive voltage applied to the three-phase coil is updated according to the PWM period and the drive voltage is applied by the output unit, so that the excitation position can be detected.

[0045] When the drive voltage varies, assuming the inductance of the motor is Lm, the resistance value is Rm, the drive voltage is V, and the energization time is t, the motor current Im satisfies the following equation: [Mathematical formula 1] The MPU 51 substitutes the energization time t1 obtained according to the specified drive voltage V1, the period of the PWM, and the specified duty ratio into (Equation 1) to obtain the motor current Im. Solving (Equation 1) for t gives: [Mathematical formula 2] t = (Lm / Rm)log e (-V / ImRm - V)...(Equation 2) Substitute the motor current Im obtained from (Equation 1) and the drive voltage V2 after the drive voltage V1 changes into (Equation 2), calculate the energization time t2, divide this energization time t2 by the PWM period, calculate the duty ratio, and the MPU 51 applies the drive voltage V2 with the updated duty ratio through the inverter circuit 52 to perform low-speed operation.

[0046] For example, in Figure 10 the case where the duty ratio is fixed at 20% and the drive voltage changes, in the waveform diagram of the induced voltage induced in the non-energized phase coil, relative to the curve graph of the drive voltage of 18V and the duty ratio of 20%, according to this motor, the duty ratio of the drive voltage of 12V obtained from (Equation 2) is approximately 32%, the duty ratio of the drive voltage of 14V is approximately 26%, the duty ratio of the drive voltage of 22V is approximately 16%, and the duty ratio of the drive voltage of 24V is approximately 15%. In Figure 10 the example, the drive voltage is set to 18V because in the motor shown in Figure 10 the optimal duty ratio (energization time) is obtained by setting the drive voltage to 18V. The duty ratio is set to 20% because in the motor shown in Figure 10 the changes in the positive and negative induced voltages generated in the non-energized phase are significant and easy to measure, and it is not a limitation on the drive voltage and duty ratio of the present invention. Figure 11 It shows the waveform diagram of the induced voltage induced in the non-energized phase coil when the duty ratio is corrected according to these voltage changes. According to Figure 11 the curve graph, it can be seen that if the duty ratio is corrected according to the change in the drive voltage, the range of voltage change is small regardless of the drive voltage, thus improving the controllability.

[0047] Thus, when the driving voltage fluctuates during low-speed operation with a driving voltage applied to the three-phase coil at a specified duty ratio, the following equation holds: [Mathematical formula 1] The MPU 51 substitutes the energization time t1 obtained based on the specified driving voltage V1, the period of the PWM, and the specified duty ratio into (Equation 1) to obtain the motor current I m , and by solving (Equation 1) for t, we get: [Mathematical formula 2] t = (Lm / Rm)log e (-V / ImRm - V)...(Equation 2) The MPU 51 substitutes the motor current Im obtained from (Equation 1) and the driving voltage V2 after the driving voltage V1 fluctuates into (Equation 2), calculates the energization time t2, divides the energization time t2 by the PWM period, calculates and updates the duty ratio, and applies the driving voltage V2 at the updated duty ratio through the inverter circuit 52. The ADC 55 measures the energized phase voltage and the non-energized phase voltage during the on-cycle of the PWM energization to detect the excitation position of the motor. Thus, it is possible to detect the excitation position of the motor without being affected by the voltage fluctuation of the driving voltage and perform low-speed operation.

Claims

1. A method for detecting the excitation position of an electric motor, the electric motor used includes: A rotor with permanent magnet excitation; a stator with three-phase coils; An output unit that supplies power to the three-phase coils bidirectionally via a half-bridge inverter circuit; a control unit that performs PWM control on the coil output according to an instruction from an upper controller, stores energization angle information and energization direction information in units of 60 energization intervals that can rotate continuously, and performs switching control on the output unit based on this information to switch the energization state; and a measurement unit that performs A / D conversion on the three-phase coil voltage and sends it to the control unit, The control unit periodically performs 120° energization including a cut-off cycle in the energization direction where the position where self-excitation stops by two-phase fixed energization coincides with the starting position of the 60° energization interval via the output unit, measures the energized phase voltage and non-energized phase voltage during the conduction cycle of PWM energization through the measurement unit, and operates in a sensorless manner while detecting the excitation position of the motor. The feature of the excitation position detection method of this motor is that When the drive voltage varies during operation with a drive voltage of a specified duty ratio applied to the three-phase coils, When the inductance of the motor is Lm, the resistance value is Rm, the drive voltage is V, the energization time is t, and the motor current is Im, the following formula holds: [Mathematical formula 1] The control unit substitutes the energization time t1 obtained from a specified drive voltage V1, the period of PWM, and a specified duty ratio into (Formula 1), obtains the motor current Im, and solves (Formula 1) for t to get: [Mathematical formula 2] t = (Lm / Rm)loge(-V / ImRm - V)...(Formula 2) Substitute the motor current Im obtained from (Formula 1) and the drive voltage V2 after the drive voltage V1 changes into (Formula 2), calculate the energization time t2, divide this energization time t2 by the PWM period, calculate and update the duty ratio, and the control unit operates by applying the drive voltage V2 with the updated duty ratio through the output unit.

Images