JP2011024401A - Starting method and driving method for brushless motor, and driving device for the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To attain stable start with high starting torque and accelerating performance according to load fluctuations, and to always achieve control without out-of-step while having quick speed followability at rapid acceleration and rapid deceleration after start. <P>SOLUTION: A counter-electromotive voltage induced in a stator winding along with the rotation of a rotor is controlled by detecting the counter-electromotive voltage in a position to provide commutation timing to determine a commutation timing. The commutation timing is detected at the phase where the counter-electromotive voltage turns positive and at the phase where the counter-electromotive voltage turns negative, and the timing is treated as the commutation timing. The commutation timing is measured in an acceleration state after start and at a position having a large amplitude of the counter-electromotive voltage, the commutation timing is measured in a large counter-electromotive voltage state even at start so that flexible start under various load conditions is realized through commutation control by measuring voltage. Out-of-step at rapid acceleration and rapid deceleration after start is also prevented. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a starting method and a driving method of a brushless motor, and more particularly to a starting method and a driving method of a brushless motor, and an apparatus thereof, which can optimally control the commutation timing of a stator by a sensorless starting and driving method.

  Conventionally, this type of sensorless driving method is, for example, as described in Japanese Patent No. 2722750, a position where the magnetic pole of the rotor changes on a stator winding called a zero cross point from a counter electromotive voltage generated in the stator winding of a motor during operation. To measure. Since the zero cross point is a stationary position that does not depend on the rotor speed, this position is measured, and from this position, the optimum commutation timing position where the rotor has rotated by an electrical angle of 30 degrees is estimated from the rotor rotational speed. The method of commutation is taken. Further, at the time of start-up, as a synchronous motor or a stepping motor, a method of forcibly commutating at a preset frequency and voltage and gradually accelerating while maintaining a load and balance until a rotation region where a back electromotive voltage is sufficiently generated is generally used. .

Japanese Patent No. 2722750

  Since the back electromotive force is small and unstable at the initial stage of the start, pre-programmed forced commutation is performed, so that it cannot be flexibly handled in an application in which the load condition changes at the start. In addition, if the electrical angle changes abruptly due to the rapid acceleration or deceleration of the rotor even after starting, the estimated error of the timing corresponding to the electrical angle of 30 ° due to the measurement of the rotational speed becomes large, and it is removed in the worst case. There was a problem of reaching the key. For example, a drive motor for an electric vehicle requires a stable start under various load conditions such as starting on flat ground, uphill, downhill, and load, and even after starting, it has rapid acceleration / deceleration performance to avoid danger. is necessary. There should be no risk of step-out that compromises safety. Thus, application to electric vehicles is the most difficult area.

  The present invention has been made in view of the above points. The first object is to realize a stable start-up with high starting torque and acceleration performance according to load fluctuations, and the second object is a quick start after the start. The aim is to realize high-trackability with acceleration / deceleration and control without step-out in any case.

Claim 1 of the present invention is characterized in that commutation control is performed by measuring a counter electromotive voltage induced in a stator winding in accordance with rotation of a rotor at a position where direct commutation timing is provided. There are two commutation timings when the rotor magnetic pole changes from the S magnetic pole to the N magnetic pole, and when the rotor magnetic pole changes from the N magnetic pole to the S magnetic pole, and are alternately generated according to the rotation direction of the rotor. The phase where the back electromotive force voltage changes to positive is the P phase, the phase where it changes negative is the N phase, and the respective commutation determination voltages are the P phase commutation determination voltage and the N phase commutation inversion voltage. The counter electromotive voltage induced according to the rotation of the rotor is measured, and the commutation occurs when the counter electromotive voltage is equal to or higher than the P phase commutation determination voltage or equal to or lower than the N phase commutation determination voltage.

  Although this is shown in FIG. 1, since the commutation timing position is directly detected instead of measuring and estimating the zero cross point as in the prior art, the delay time to the commutation timing is estimated as in the case of the zero cross point detection. There is no need, and reliable commutation can be realized with almost no danger of step-out. In addition, since measurement can be performed at a position where the back electromotive voltage has a large amplitude, particularly effective control can be performed in a speed region where the back electromotive voltage is small.

Claim 2 is a means made possible by measuring the commutation timing at a position where the back electromotive voltage is large as described above, and performing commutation control with reference to the back electromotive voltage from the first driving of starting. This is a characteristic starting method. Generally, the optimum commutation timing is at a position about 30 ° in electrical angle from the position of the zero cross point, and is at a value of about 50% of the maximum amplitude of the back electromotive force. Further, as will be described later, if the discontinuity of the generated torque obtained from the drive current is allowed, the back electromotive voltage for measuring the commutation timing can be allowed up to the vicinity of the maximum value of the amplitude. Therefore, the commutation timing can be detected with a sufficiently large counter electromotive voltage even at the time of starting.
The commutation determination voltage must be smaller than the counter electromotive voltage induced by the starting driving force, but this condition is also to provide an initial driving power that generates an electromotive voltage large enough to be detected. Then, after giving a large initial driving force within a range suitable for each application, the optimum value of the commutation determination voltage is experimentally determined in consideration of load fluctuations.

  According to the third aspect of the present invention, a commutation determination voltage is determined in accordance with the rotational speed of the rotor, and the commutation determination voltage is determined. The back electromotive force is a function of the electrical angle and the rotor rotational speed, and further varies depending on the motor structure. Most simply, the value of the counter electromotive voltage at the commutation timing at an electrical angle of 30 ° is calculated from the counter electromotive voltage constant defined by each motor and used as the commutation determination voltage. Actually, it is affected by the motor structure and the operation delay of the energization control circuit, etc., so select the representative speed at several points and measure the back electromotive force at the commutation timing and interpolate to obtain it. In the range, it is preferable to measure and set the representative value. In applications where the speed changes frequently, the latter measure is substantial. The driving device of this driving method is claim 6, and details thereof will be described in an embodiment described later.

In order to facilitate understanding of the present invention, basic motor operation will be described. FIG. 2 shows the configuration of a general brushless motor drive circuit, in which current is applied to the U phase, V phase, and W phase of the stator via an inverter circuit. The position detection circuit detects the rotor zero-cross point with reference to the back electromotive voltage from the terminal voltage of each phase, and the rotor speed is counted by the speed counting circuit to determine the commutation timing. FIG. 3 shows the commutation timing and the energization step, and the horizontal axis represents the rotation of the rotor in electrical angle. For energization of each phase, steps (1) to (6) are repeated in the order of U phase → V phase, U phase → W phase, and V phase → W phase as shown in (d). The counter electromotive voltage induced in each phase is as shown in (e), where the electrical angle θ is the relative position between the position where the magnetic pole of the rotor switches and the U phase winding of the stator is the electrical angle θ. Can be approximated by the following equation.
Eu 回 転 Rotational speed x SINθ
The same applies to the V phase and the W phase, and the phases are different by 120 °. θ = 0 is a position where the magnetic pole of the rotor switches on the U-phase winding, and the back electromotive voltage crosses the zero potential here, so it is called a zero cross point.

The torque generated by energization can be expressed as follows, for example, in the U-phase → V-phase energization.
Tuv ∝ Energizing current x COS (θ-60 °)
If the commutation is performed at a position of 30 ° before and after θ = 60 ° at which the torque is maximized, driving with maximum efficiency can be obtained. That is, an electrical angle of 30 ° is an ideal commutation timing point. When ideal commutation control is performed in this way, the maximum efficiency torque line is obtained as shown in FIG. Actually, as shown in (f) of FIG. 3, the terminal voltage of each phase is, for example, in the commutation step (6), the W → V phase is energized and the U phase has no energization current. The back electromotive force of (e) is superimposed at the center. A position where this voltage crosses Vd / 2 is detected and set as a zero cross point. Based on this position, the electrical angle lead angle of 30 degrees is estimated from the rotor rotational speed, and the commutation step (1) is executed.

  The above is the control when the motor is in steady rotation. At the time of starting, the rotor starts from zero speed, so the counter electromotive voltage near the zero cross point is very small. In particular, it cannot be distinguished whether it is a zero potential indicating a zero cross point or a zero potential generated due to a low speed. Moreover, at the time of starting, it may start from the position after the zero cross point. Even if the zero-cross point can be detected, it is difficult to estimate the timing corresponding to the electrical angle of 30 ° because the rotational speed of the rotor is indefinite. These causes are the reason why the conventional starting method has to rely on the forced commutation of the program method.

On the other hand, the present invention measures the back electromotive force at a position where a significant back electromotive force is generated by accelerating the rotor by moving more than 30 degrees in electrical angle from the stop position without detecting the zero cross point. Since the flow timing is determined, the causes of the conventional method described above are solved at once. The behavior of the back electromotive force when the engine is in the acceleration state from the stop at the beginning of the start can be explained as follows. If the angular velocity of rotation is ω in a steady state, back electromotive force ∝ ω × SINωt ωt = θ (electrical angle)
Since ω is proportional to time in the acceleration state at the time of starting, ω → ω ₀ t, ωt → ω ₀ t ² = θ, and further, when starting from this position with the electrical angle of the rotor stop position being θ _x It can be expressed as power ∝√ (ω ₀ θ) × SIN (θ + θ _x ).
Here, in the range where θ + θ _x is small, that is, the range in which the sine function can be approximated by a linear line, back electromotive force ∝√ (ω ₀ θ) × (θ + θ _x )
= √ω ₀ × (θ ^1.5 + θ _x θ ^0.5 )

FIG. 4 illustrates this result. When the rotor starts from θ _x1 before the zero cross point, the back electromotive voltage swings from zero to minus and passes through the zero cross position, but this time is masked to selectively select the zero cross point. It is almost impossible to detect. Of course the zero-crossing point, if you start from θ _x2 can not be detected without. On the other hand, in both cases, the back electromotive force after the zero cross point is steep and is steep and increases by the power of 1.5 as the electrical angle increases. It can be seen that a significant counter electromotive voltage can be expected by accelerating to 30 ° by the electrical angle at the previous commutation timing.
Even when the starting position is not in the vicinity of the above-mentioned zero cross but further deviated, as described in FIG. 5 below, the commutation timing can be allowed up to a larger electrical angle. It is possible to detect the commutation timing at a position where is increased. That is, it is possible to reliably grasp the back electromotive voltage from the start energization, enabling commutation control using the back electromotive voltage from the start, and realizing a stably controlled start acceleration.

FIG. 5 shows the case of commutation at point A, deviating from the ideal commutation timing due to the relationship between the back electromotive force and the generated torque. Such a deviation is typical when the rotor stop position measurement before starting is insufficient, when the estimation of the rotational speed of the rotor at the time of starting is not appropriate, or when sudden acceleration or sudden deceleration occurs after starting To happen. Here, it is assumed that the commutation timing deviates for some reason and deviates by 30 ° or more in electrical angle from the ideal commutation point and comes to point A. Even if an attempt is made to detect the zero cross point in the W phase by switching from the energization of the W phase to the V phase and the U phase to the V phase to detect the zero cross point, the point A has already passed the zero cross point of the W phase. In the method, the zero cross point cannot be detected, the commutation timing is missed, and the _TUV is continued as it is, resulting in a negative torque, that is, a step-out phenomenon. On the other hand, in the present invention, a state where the zero cross point is passed and the state is changed to negative as shown in the figure is detected, and even if the state is changed to negative and is shifted to the vicinity of the electrical angle of 90 ° at the negative peak position, it is detected as the commutation timing. . Since _TUV is a forward torque during this period, acceleration of forward commutation is continued without causing step-out. That is, even if there is a large change in the electrical angle for some reason, as long as the back electromotive voltage exceeds the commutation determination voltage, torque discontinuity will occur but forward acceleration will be avoided to avoid the risk of step-out. Can keep. This also absorbs the error of the initial position of the rotor at the time of starting, and further increasing the initial driving force is a direction advantageous for detecting the commutation timing by further increasing the back electromotive voltage. Make it possible to give. Even if there is sudden acceleration / deceleration in the driving state after starting, as long as the back electromotive voltage exceeding the commutation judgment voltage is induced, there is almost no risk of step-out, and it can be controlled by an electric vehicle that requires safety. Can be applied optimally.

An embodiment in which the present invention is applied to an electric motorcycle will be described with reference to the functional block diagram of FIG. Reference numeral 1 denotes a 48V battery that supplies power to a brushless motor (hereinafter referred to as a motor) 3 through an inverter circuit 2. The motor 3 is a direct drive system in which the stator winding is 51 poles, the rotor magnetic pole is 48 poles and 24 pole pairs, the stator is fixed to the axle, and the rotor is a part of the wheel. At a rated voltage of 48 V, a rated current of 20 A, the maximum rotation speed is 400 RPM, and the counter electromotive voltage coefficient is 0.1 V / RPM. The wheel diameter is 50 cm, which corresponds to a movement of about 1.5 m per rotation of the wheel, and is 36 RPM per hour at 400 RPM. Reference numeral 5 denotes a driver circuit for a switching element of the inverter circuit.

6 receives an operation command signal 8 for starting, acceleration and deceleration from the outside and energizes the back electromotive force signal 7 obtained by resistance-dividing the terminal voltage of the three-phase winding of the motor and the current detector 4 including a shunt resistor. This is an energization control circuit that inputs a current signal and controls on / off of the switching element of the inverter circuit 2. The energization control circuit 6 is composed of an 8-bit microcomputer with a built-in motor control circuit having an operating frequency of 16 MHz. Since the driving voltage of the motor is 48 volts, the terminal voltage of the winding is resistance-divided into 1/10 and input to the energization control circuit 6 as a back electromotive voltage signal 8. Therefore, the counter electromotive voltage signal at commutation timing detection is
V = back electromotive voltage + 2.4V However, when pulse width modulation signal is on (1)
Is superimposed on 2.4V.

  The operation command signal 8 is input to the operation control circuit 6a, the start command is sent to the rotor initial position detection circuit 6b, the acceleration / deceleration command is converted into a duty ratio command signal, and sent to the pulse width modulation circuit 6e. The rotor initial position detection circuit 6b to which the start command is given gives a command to the energized phase distribution circuit 6c to energize each winding with a short pulse that does not generate a driving torque to the motor, and the current detector 4 synchronizes with the energization. The energizing current signal is measured, the rotor position is determined from the magnitude of the energizing current of each winding, and the starting energized phase is determined. This information becomes a start energization start command, is returned to the energization phase distribution circuit 6c, and energization of the motor 3 is started via the pulse width modulation circuit 6e, the driver circuit 5, and the inverter circuit 2. The energized phase distribution circuit 6c simultaneously supplies energized phase information and energization timing signal to the commutation timing determination circuit 6d, and detects the commutation timing to the next energized phase.

The commutation timing determination circuit 6d is controlled by the timing signal T having the pulse width modulation signal from the pulse width modulation circuit 6e, the energization timing and energization phase information from the energization phase distribution circuit 6c, and receives the back electromotive voltage signal 7. The phase selection circuit 6d1 for selecting the phase for detecting the commutation timing, the commutation determination voltage generation circuit 6d3 for generating the commutation determination voltage used for determining the commutation timing of the selected phase, and both of these signals. The comparison circuit 6d2 is inputted and determines the commutation timing. 6d4 is a speed counting circuit for measuring the timing signal T. The 6d3 commutation determination voltage generation circuit has a P phase commutation determination voltage and an N phase commutation determination voltage determined by the rotational speed of 6d4 as a table, and switches and outputs them. Since the comparison circuit 6d2 refers to the value shown in the equation (1), the commutation timing is detected in synchronization with the ON timing of the pulse width modulation signal, that is, the state where the motor is energized.
In this way, when the commutation timing determination circuit 6d detects the next commutation timing, this is sent to the energized phase distribution circuit 6c, and energization of the next commutation phase is started.
The table below is a table of commutation determination voltages used in this example.

FIG. 7 shows an example of the operation of this embodiment, in which the U-phase, V-phase, and W-phase terminal voltages at the start are measured. (1) and (2) are measurement examples in the case of commutation from the P phase and commutation from the N phase at the time of no-load start when no person is on board. The short pulses observed during the first approximately 10 ms of the start are those when the rotor position is detected from the magnitude of the current value by passing a current through each phase. In (1), the U phase → W phase is selected, and in (2), the U phase → V phase is selected, and the start-up energization has begun. The counter electromotive voltage is observed with 24V as the virtual midpoint for the V phase and the W phase, respectively. The former is the phase in which the counter electromotive voltage increases, and the latter is the phase in which the counter electromotive voltage decreases. It has moved to the commutation step. Starting energization is performed for (1) for about 13 ms and (2) for about 6 ms. This difference is considered to be because the rotor stop position at the start of the two was different. {Circle over (3)} is an example of measurement in the case of a high load start where a person gets on. After the rotor position is detected, the start-up energization is performed from the W phase to the U phase, the counter electromotive voltage is observed in the V phase, and the commutation timing is detected. In this case, it takes about 40 ms from the start energization to the first commutation because of the high load start. As is clear from these examples, external fluctuation factors at the start, such as differences in the initial stop position and load magnitude, are reflected in the commutation timing and absorbed. It can be seen that an extremely smooth start can be realized.

Illustration of commutation judgment voltage and commutation timing Operation explanatory diagram of brushless motor drive device Illustration of signal waveform for driving brushless motor Illustration of back electromotive force behavior at start-up Illustration of deviation of commutation timing The block diagram which shows the Example of this invention Explanatory diagram of operation waveforms of the embodiment

1. Battery 2. Inverter circuit 3. 3. Brushless DC motor 4. Current detection circuit Driver circuit 6. Energization control circuit 6a. Operation control circuit 6b. Rotor initial position detection circuit 6c. Energized phase distribution circuit 6d. Commutation timing determination circuit 6d1. Phase selection circuit 6d2. Comparison circuit 6d3. Commutation determination voltage generation circuit 6d4. Speed counting circuit 6c. 6. Pulse width modulation circuit 7. Back electromotive voltage signal Operation command signal

Claims (6)

  In a method of driving a brushless motor connected via an inverter circuit formed by bridge-connecting a plurality of switching elements to a DC power source by the energization control of the inverter circuit, the energization control is a counter electromotive force that is induced in a stator winding. A P phase commutation determination voltage in a direction in which the voltage increases beyond the zero cross point potential and an N phase commutation determination voltage in a direction in which the voltage decreases beyond the zero cross point potential. The plurality of switching elements are sequentially switched and controlled with a commutation timing that is a time point when the phase commutation target voltage is reached or more and a back electromotive voltage generated by the rotation of the rotor reaches an N phase commutation determination voltage or less. A method for driving a brushless motor.   In the brushless motor start method, the energization control determines the energization phase of the start from the measurement of the rotor stop position, and the back electromotive voltage induced by energizing the start power is equal to or higher than the P phase commutation determination voltage. 2. The method of starting a brushless motor according to claim 1, wherein the starting power is commutated by detecting a time point when the voltage reaches the N phase commutation determination voltage or less. The energization control is performed in advance at the commutation timing position corresponding to 30 degrees in electrical angle with respect to the stator winding in the direction in which the counter electromotive voltage increases and decreases in each rotation speed of the rotor. The value of the counter electromotive voltage in the direction in which the counter electromotive voltage increases is the speed-dependent P phase commutation determination voltage, and the value in the direction in which the counter electromotive voltage decreases is the speed dependent N phase commutation determination voltage, Further, the minimum speed at which the speed-dependent P-phase commutation determination voltage and the speed-dependent N-phase commutation determination voltage at the rotor speed below the limit of detection of the counter electromotive voltage of the low speed or the low speed can be significantly measured by the above measurement, respectively. As a value obtained in the above, the time when the back electromotive voltage reaches the speed dependent P phase commutation target voltage or more, or the time when the back electromotive voltage reaches the speed dependent N phase commutation determination voltage or less is used as the commutation timing. of Claim 1 and a driving method of a brushless motor according to claim 2, characterized in that so as to sequentially switch controls the switching element. In the speed-dependent energization control, the speed-dependent P-phase commutation determination voltage and the speed-dependent N-phase commutation inversion voltage are average values of arbitrary speed control regions. 4. The method of driving a brushless motor according to claim 3, wherein the commutation timing is detected and determined by a P-phase commutation determination voltage and a speed-dependent N-phase inversion voltage. An inverter circuit formed by connecting a plurality of switching elements to a DC power supply, a brushless motor having a stator winding connected to the output end thereof, and a counter electromotive force that is connected to the stator winding and induced in the winding is input. In the brushless motor driving device comprising an energization control circuit that controls on / off of a plurality of switching elements of the inverter circuit, the driving device generates a P-phase commutation determination voltage and an N-phase commutation determination voltage. The P-phase commutation determination voltage and the N-phase commutation determination voltage connected to the output terminal in synchronization with the commutation timing in the order determined by the rotation direction of the rotor and the energization direction of the stator winding. Is connected to the output terminal and the stator winding, and the counter electromotive voltage induced in the stator winding is A commutation judgment comparison circuit for sending a commutation command when the P phase commutation judgment voltage is reached or below the N phase commutation judgment voltage, and the plurality of switching elements are sequentially switched by the commutation command. A brushless motor drive device characterized in that it is configured. The energization control circuit includes a speed counting circuit that counts the commutation timing signal and measures the rotational speed of the rotor, and the commutation determination voltage generation circuit is connected to an output of the speed counting circuit, and from the rotor rotational speed, 6. The brushless motor according to claim 5, wherein said speed-dependent P-phase commutation determination voltage and said speed-dependent N-phase commutation determination voltage are output to said commutation determination voltage output circuit as means for outputting. Drive device.

Images