JP5965517B2 - Motor control device and control device - Google Patents

Description

  The present invention relates to a motor drive control device.

  For example, a motor-assisted bicycle equipped with a motor has a mechanism that uses a sensor (such as a torque sensor) that detects human power, and adjusts (assist control) the electric assist force by the motor in accordance with the detection signal.

  In addition, in order to be able to travel as long as possible with the energy of the battery installed, the motor is operated as a generator during deceleration, such as during braking, energy recovery (regeneration) is performed, and the electric power generated by the motor is used. I try to charge the battery.

  In such an electrically assisted bicycle, motor drive control is performed in consideration of not only the target torque obtained from a detection signal of a torque sensor or the like but also the current rotational speed of the motor.

  In general, the rotational speed of a motor is detected by a method of detecting the period of a pulse signal (Hall signal) output from a rotation sensor such as a Hall element and calculating the reciprocal thereof. However, since the past pulse interval is essentially measured, the current rotational speed cannot be obtained with sufficient accuracy.

  For example, in a bicycle with auxiliary power described in JP-A-10-258788, a pulse interval output from a rotation sensor is counted by a timer, and this is converted into a speed, and the converted speed is converted into a predetermined time. A driving force corresponding to the averaged speed is determined. That is, the converted speed is averaged over a predetermined time, and the driving force corresponding to the averaged speed is determined.

Japanese Patent No. 2623419 JP 2003-276672 A Japanese Patent No. 3317096 Japanese Patent Laid-Open No. 10-258788

  Accordingly, an object of the present invention is to provide a technique for making it possible to accurately estimate the current speed used in motor drive control.

  Another object of the present invention is to provide a technique for enabling detection of a malfunction in a rotation sensor of a motor used when estimating a current speed.

  The motor drive control device according to the first aspect of the present invention receives (A) a periodic signal that is received from a rotation sensor of a motor and can be divided into a plurality of phases according to the rotation phase of the motor. A measuring unit that measures time for each of the plurality of phases, and (B) a time corresponding to one period of the signal based on at least a part of the time in the phase in the predetermined observation section, measured by the measuring unit. And a speed calculation unit that estimates a current speed from the sum and a value according to the amount of change in time of a phase in a predetermined observation section, and drives the motor according to the estimated current speed. A control unit for controlling.

  In this way, the period is accurately calculated by adjusting the time for one period according to the amount of change in the time of a phase among a plurality of phases (for example, partial periods in the embodiment) in a predetermined observation section. As a result, the current speed can be calculated with high accuracy.

  Further, the measurement unit described above may specify the rotation direction for each of a plurality of phases from the signal described above. In this case, the speed calculation unit described above, when the rotation direction is constant in the predetermined observation section and each time of the plurality of phases is less than the predetermined time in the predetermined observation section, the first calculation in the observation section The amount of change may be calculated based on the difference between the phase time and the second phase time.

  During normal driving, it is preferable to perform acceleration compensation by such calculation. Note that the predetermined coefficient is determined from the detection delay, the control delay, and the like for the first phase and the second phase. Further, a delay corresponding to the frame latency for one period may be considered. It may add a fixed delay.

  The predetermined observation section described above may include a plurality of phases for one cycle of the signal and a phase one cycle before the signal corresponding to the latest phase among the plurality of phases. In this case, the amount of change described above may be calculated by the difference between the time of the latest phase and the time of the phase one cycle before the signal corresponding to the latest phase.

  Moreover, the measurement part described above may specify a rotation direction about each of several phases from the said signal. In this case, the speed calculation unit described above, when there is a phase whose time has exceeded a predetermined time or whose rotation direction is reversed, in addition to the most recent predetermined number of phases in the predetermined observation section, The time corresponding to one cycle of the signal may be calculated from one of the phases, and the amount of change may be calculated from the time difference between the two phases included in the most recent predetermined number of phases. As described above, when the rotation direction is slightly reversed or a little after the start of traveling, the speed cannot be calculated as in normal traveling, so the speed is estimated from the time of the most recent predetermined number of phases.

  Note that the most recent predetermined number described above is 2, and the time corresponding to one cycle of the signal may be calculated from the latest phase among the most recent predetermined number of phases.

  Furthermore, the measurement unit described above may specify the rotation direction for each of a plurality of phases from the signal. In that case, the speed calculation unit described above has the reversal of the rotation direction in the most recent predetermined number of phases in the predetermined observation section, or at least one of the most recent predetermined number of phases is equal to or longer than the predetermined time. In this case, the time corresponding to one cycle of the signal may be calculated from the latest phase, and the current speed may be estimated from the time corresponding to one cycle of the signal. When the reversal is just started or when the speed is low, it may be dealt with in this way.

  The control device according to the second aspect of the present invention receives (A) a periodic signal received from a rotation sensor of a motor and can be divided into a plurality of phases according to the rotation phase of the motor. And (B) an average time for one phase obtained from the time for one cycle of the signal using the time for each of the plurality of phases measured by the measuring unit. A rotation sensor when a predetermined condition is satisfied by comparing the average or weighted average of the time of a phase of the plurality of phases with the time of the corresponding phase one cycle before the signal from the phase And a determination unit that outputs the error.

  By doing so, it becomes possible to accurately detect an error of the rotation sensor.

  The determination unit described above may determine whether the ratio between the average time and the average or the weighted average is within a predetermined range.

  Further, the determination unit described above may output an error when a state satisfying a predetermined condition continues for a predetermined time. This is because there may be an error state temporarily.

  Note that the control device according to the second aspect may further include a control unit that controls driving of the motor in accordance with the error.

  The embodiment described below is merely an example, and there are various modes for realizing the configuration described above. In some cases, it is realized not only by hardware but also by a combination of software and a processor.

  According to one aspect, it is possible to accurately estimate the current speed used in motor drive control.

  According to another aspect, it is possible to detect a malfunction in the rotation sensor of the motor used when estimating the current speed.

FIG. 1 is an external view of an electrically assisted bicycle. FIG. 2 is a diagram illustrating a configuration example of the motor drive control device. FIG. 3 is a diagram illustrating a configuration example of the control unit. FIG. 4 is a diagram for explaining motor driving using the target torque and the current vehicle speed. FIG. 5 is a diagram illustrating a configuration example of the vehicle speed calculation unit. FIG. 6 is a diagram for explaining the hall signal. FIG. 7 is a diagram illustrating a configuration example of the partial cycle detection unit. FIG. 8 is a transition diagram of signals generated by the partial period detection unit. FIG. 9 is a diagram for explaining the observation interval. FIG. 10 is a diagram for explaining the operation mode. FIG. 11 is a diagram illustrating a configuration example of a control unit according to the second embodiment.

[Embodiment 1]
FIG. 1 is a diagram showing an external appearance of an electrically assisted bicycle according to a first embodiment of the present invention. The electric assist bicycle 1 is of a general rear wheel drive type in which a crankshaft and a rear wheel are connected via a chain, and includes a secondary battery 101, a motor drive control device 102, a torque sensor 103, a brake sensor. 104, an operation panel 105, and a motor 106 are provided.

  The secondary battery 101 is a lithium secondary battery whose supply maximum voltage (voltage at full charge) is 24 V, for example, but other types of secondary batteries may be used.

  The motor drive control device 102 is accommodated in, for example, a housing fixed to the upper part of the front wheel. Details of the motor drive control device 102 will be described later in detail.

  The torque sensor 103 is provided on the crankshaft, detects the pedaling force applied by the passenger, and outputs the detection result to the motor drive control device 102.

  The brake sensor 104 includes, for example, a magnet and a known reed switch. The magnet fixes the brake lever and is fixed to the brake wire connected to the brake lever in the casing through which the brake wire is passed, and turns on the reed switch when the brake lever is gripped. The reed switch is fixed in the casing. A reed switch conduction signal is output to the motor drive control device 102.

  The operation panel 105 is fixed at the center of the handle, and instruction information such as the start of assist is output from the operation panel to the motor drive control device 102.

  The motor 106 is, for example, a well-known three-phase DC brushless motor. The motor 106 is attached to, for example, the front wheel of the electric assist bicycle 1 and rotates the front wheel, and the rotor rotates so that the rotor rotates according to the rotation of the front wheel. It is connected to. Further, the motor 106 includes a Hall element and the like, and outputs rotation information of the rotor to the motor drive control device 102.

  Next, a circuit example of the motor drive control device 102 and the like will be described with reference to FIG.

  The motor drive control device 102 includes a control unit 300 and a drive unit 400. The control unit 300 is connected to the torque sensor 103, the brake sensor 104, and the operation panel 105. Furthermore, Hall signals (Hall-U, Hall-V, Hall-W) from Hall elements of the motor 106 are input to the control unit 300 and the driving unit 400.

  The control unit 300 may be implemented by a dedicated circuit, or may be implemented by a combination of a program for executing the processing described below and a processor that executes the program. It may be implemented by a combination of a circuit and a processor and a program. When implemented by a combination of a processor and a program, the program is held in a nonvolatile memory, and the processor reads the program into a RAM (Random Access Memory) and executes it. Further, fixed data used for executing the program is stored in a nonvolatile memory.

  Further, when an instruction to start assist is given from the operation panel 105 and a signal corresponding to the input torque output from the torque sensor 103 is input, the control unit 300 outputs a control signal corresponding to the current vehicle speed and the target torque. Output to 400. Note that a hall signal included in the motor 106 is used for this processing.

  When a brake operation is performed and a brake signal is input from the brake sensor 104 to the control unit 300, the control unit 300 converts the output voltage of the drive unit 400 to the motor electromotive force based on the input signal from the brake sensor 104. The brake control force can be obtained by regenerating electric power from the motor 106 under the following control. Such regenerative processing is not a main part of the present embodiment, and thus further explanation is omitted.

  The drive unit 400 includes a drive control circuit 410 that performs drive control from a control signal from the control unit 300 and a hall signal, a drive circuit 420 that includes drive elements 421 to 423, and FETs (Field Effect Transistors) Q11 to Q32. Including a three-phase bridge circuit.

  In the three-phase bridge circuit, FETs Q11 and Q12 are connected in series for the U phase, FETs Q21 and Q22 are connected in series for the V phase, and FETs Q31 and Q32 are connected in series for the W phase. . In response to a control signal from the control unit 300, the drive control circuit 410 applies a PWM (Pulse Width Modulation) signal (U-PWM, V-PWM, W-PWM) to each of the three drive elements 421 to 423 in the drive circuit 420. ) To control the switching operation of each FET. Then, the ratio (duty ratio) of the application time to the motor 106 and the non-application time of the voltage Vbat applied from the secondary battery 101 is adjusted. In this manner, the power supplied to the motor 106 (Vu, Vv, Vw) is adjusted to achieve an appropriate rotation speed of the motor 106. Since a specific driving method of the FETs Q11 to Q32 is well known, description thereof is omitted here.

  Next, a configuration example of the control unit 300 will be described with reference to FIG. The control unit 300 includes a target torque calculation unit 350, a vehicle speed calculation unit 310, a first duty ratio conversion unit 330, a second duty ratio conversion unit 320, and an addition unit 340.

  The target torque calculation unit 350 calculates the target torque from a signal corresponding to the input torque from the torque sensor 103 and outputs the target torque to the first duty ratio conversion unit 330. The first duty ratio conversion unit 330 performs processing such as multiplication by a conversion coefficient (duty ratio / coefficient corresponding to torque), calculates a first value converted into the duty ratio, and outputs the first value to the addition unit 340. . On the other hand, the vehicle speed calculation unit 310 calculates the current vehicle speed from the hall signal as described in detail below, and outputs it to the second duty ratio conversion unit 320. The second duty ratio conversion unit 320 performs processing such as multiplying the current vehicle speed by a conversion coefficient (duty ratio / coefficient corresponding to the current vehicle speed) and the like, and adds the second value converted into the duty ratio to the addition unit 340. Output to.

  Adder 340 adds the first value from first duty ratio converter 330 and the second value from second duty ratio converter 320 to generate a control signal, and outputs the control signal to driver 400. . The drive unit 400 drives the motor 106 in accordance with the control signal as described above.

The operation of the motor drive control apparatus 102 will be further described with reference to FIG. Here, it is assumed that the current vehicle speed is represented as D ₀ and the target torque is represented as D _T. Then, as described above, the average duty ratio Duty of the FET in the three-phase bridge circuit is expressed as follows.
Duty = D ₀ + D _T

In the present embodiment, the average duty ratio Duty is changed along the straight line D ₀ if the current speed is maintained. When accelerating the motor 106 in the power running state, the target torque is set to a positive value, for example, + D _t , so this straight line is shifted upward by D _t . If it does so, it will accelerate by the amount of target torque, but an average duty ratio will also be made relatively high. On the other hand, when decelerated by the motor 106, for example, in the braking state, the target torque negative value, for example, because it is set to -D _t, is shifted to the straight line downward D _t. Then, the speed is reduced by the target torque, and the average duty ratio is also relatively lowered.

Furthermore, in response to forward rotation and reverse rotation of the motor 106, the current vehicle speed D ₀ also, there is a case where positive and negative values are set.

  By performing the torque feedforward control as described above, stable control is performed, which is suitable for an electrically assisted bicycle or the like.

  Next, a configuration example of the vehicle speed calculation unit 310 will be described with reference to FIG. As shown in FIG. 5, the vehicle speed calculation unit 310 includes a partial cycle detection unit 311 and a current vehicle speed calculation unit 312. The partial cycle detection unit 311 processes the hall signal from the motor 106 as described below, and outputs the length and rotation direction of the partial cycle described below in detail to the current vehicle speed calculation unit 312. The current vehicle speed calculation unit 312 calculates the current vehicle speed according to the length of the partial cycle and the rotation direction from the partial cycle detection unit 311.

  In this embodiment, one period of the Hall signal is divided into six partial periods. As shown in FIG. 6, the Hall signals (Hall-U, Hall-V, Hall-W) are shifted from the high period to the low period. Detect and divide one period into six partial periods by these edges. However, the length of each partial period varies depending on the rotation speed of the motor 106. Further, as will be specifically described below, positive rotation (forward) and negative rotation (reverse) in the rotational direction are distinguished by the detection order of the rising edge and the falling edge.

  Next, the configuration of the partial period detection unit 311 will be described with reference to FIG. The partial cycle detector 311 includes a first-stage D-FF (flip-flop) to which each Hall signal is input and a second-stage D-FF to which the output of the first-stage D-FF is input. And a hole edge detector 3111, a phase number converter 3112 for converting the output of the first stage D-FF into a phase number (PN: Phase Number) for each period, and the output of the phase number converter 3112 A subtractor 3113 that calculates a difference between a first phase delay D-FF for delaying by one phase (that is, a partial period) and an output of the phase number converter 3112 and an output of the first phase delay D-FF. A phase edge direction determination unit 3114 that determines positive rotation or negative rotation based on an output from the subtraction unit 3113, and a second phase delay D-FF for delaying the output of the phase edge direction determination unit 3114 by one phase. And the output of the phase edge direction determination unit 3114 A partial cycle direction determination unit 3115 that determines the rotation direction from the two phase delays D-FF, a partial cycle counter 3116 that counts the number of high-frequency clocks at the interval between the hole edges output by the hole edge detection unit 3111, It has a full count detection unit 3117 that detects that the cycle counter 3116 has reached full count, and a partial cycle register 3119 that holds the output of the partial cycle counter 3116. Here, “phase” indicates a unique phase in a partial cycle unit, and the above-mentioned “phase” has the same meaning. Further, “phase” used in the following also has the same meaning.

  A high frequency clock is input to the first D-FF and the second D-FF. The high frequency clock is also input to the partial cycle counter 3116, the first and second phase delays D-FF, and the partial cycle register 3119. Also, the Hall edge detection unit 3111 calculates an exclusive OR for each of the output of the first D-FF and the output of the second D-FF related to the same Hall signal, and A hole edge is generated by calculating the sum of exclusive ORs.

  Next, the operation of the partial period detection unit 311 will be described. At the same time, see the state transition diagram of FIG.

  First, the Hall-U, Hall-V, and Hall-W signals are synchronized with the high-frequency clock by the first stage D-FF. The hall edge detection unit 3111 further delays one clock by the output of the first stage D-FF and the second stage D-FF (the clock frequency of the high frequency clock is much higher than that of the hall signal, so that the delay is delayed). A hole edge is detected by detecting a change point of the Hall signal output that is small in amount. When a logical sum is calculated for the results of detecting both rising and falling edges of the Hall-U, Hall-V, and Hall-W signals, six fine Hall edge pulses are output in one period of the Hall signal.

  The phase number conversion unit 3112 outputs a six-phase phase number PN from the combination of the HDU, HDV, and HDW signals of the first stage D-FF output. The phase numbers are 1 to 6 in order from the rising edge of the HDU during forward rotation (forward movement). During reverse rotation (reverse rotation, reverse rotation), the order is 6 to 1.

  The phase number (PN) is the phase delay phase number PND to which a delay corresponding to another phase (partial period) is added by the first phase delay D-FF. The subtracting unit 3113 subtracts the phase delay phase number PND from the phase number PN before the delay to obtain a phase number difference (PN-PND). The phase number difference is +1 or -5 (in the case of a change from phase number 6 to 1) in which the phase number increases by 1 during forward rotation (forward movement). On the other hand, during negative rotation (retreat), the phase number difference is −1 or +5 (when the phase number is changed from 1 to 6) in which the phase number decreases by 1. As a result, the next phase edge direction determination unit 3114 determines whether the previous hole edge has passed through the positive rotation or negative rotation, and outputs the edge rotation direction (1: positive rotation / 0: negative rotation). Is done.

  Further, the edge rotation direction which is the output of the phase edge direction determination unit 3114 is output to the partial cycle direction determination unit 3115 with a delay of one phase (partial period) added by the second phase delay D-FF. The The partial cycle direction determination unit 3115 compares the edge rotation directions before and after the delay. (A) If both edges before and after the delay are positive rotation, the partial cycle direction between them is also positive. (B) Both edges before and after the delay are negative rotation. Then, the partial cycle direction between them is also negatively rotated. (C) When the rotation directions are different at both edges before and after the delay, it is determined that the partial cycle direction between them is reversed (inversion occurs during the partial cycle).

  On the other hand, in the right half block of the partial cycle detector 311, the partial cycle counter 3116 is clock-synchronized cleared by the hole edge pulse, and counts up from zero with a high-frequency clock. With the next hall edge pulse, the partial cycle count value is updated and taken into the partial cycle register 3119 of the next stage, and the partial cycle register 3119 outputs the fetched partial cycle. At the same time, the partial cycle counter 3116 is synchronously cleared.

  When the partial cycle counter 3116 reaches a predetermined full count (a count value at which the vehicle speed can be regarded as almost zero) at the time of an extremely low speed that is considered to be almost stopped, the full count detection unit 3117 The count is stopped at the count enable (CE) terminal of the period counter 3116, and the count value is saturated. When the value of the saturated partial period is taken into the partial period register 3119 and output, the current vehicle speed calculation unit 312 treats it as stopped (substantially stopped).

  The signal input from the OR gate to the clock enable (EN) terminal of the partial cycle register 3119 becomes high when the partial cycle counter 3116 detects that the full cycle is detected by the full count detection unit 3117, and becomes a partial cycle register. 3119 is forcibly loaded with a full count value (saturated partial period). That is, when the full count detection unit 3117 detects a full count, the partial period is a saturated partial period.

  As a result, since the next hole edge does not come at the time of complete stop, it is possible to prevent a situation in which the last saturated partial period to be determined to be almost stopped is not taken into the partial period register 3119.

  As shown in FIG. 8, when the rotation direction is reversed in the period I, the negative rotation R is detected after one phase for the edge rotation direction, and further 1 for the phase delay edge rotation direction. The negative rotation R is detected after the phase. For the partial period rotation direction, when a negative rotation R is detected for the edge rotation direction, a reversal of the rotation direction is detected.

  Next, the calculation contents of the vehicle speed calculation unit 310 will be described.

  In the present embodiment, as shown in FIG. 9, when the current partial period is P (n), seven partial periods up to the previous partial period P (n-6) are observed, Know your driving status. That is, the latest seven partial periods and rotation directions are held in a memory such as a RAM. As shown in FIG. 9, none of the partial periods P (n) to P (n-6) is a saturated partial period (there is a speed absolute value greater than or equal to a predetermined value), and the rotational direction is the same direction (FIG. 9). (1) The current vehicle speed is calculated by performing a high-precision acceleration compensation calculation using seven partial periods.

  Further, although the latest two partial periods P (n) and P (n−1) (here, “two” may be another value as an example), the rotation is the same direction and is not a saturated partial period. If there is a reversal of the rotational direction in other parts or there is a saturated partial period, (2) the current vehicle speed is calculated by performing an acceleration compensation calculation using the latest two partial periods.

  Furthermore, in at least one of the latest two partial periods P (n) and P (n−1) (here, “two” may be another value as an example), the rotation direction is reversed. If it occurs or is a saturated partial period, (3) the current vehicle speed is calculated using only the latest partial period.

  As shown in FIG. 10, in the cases of (A) and (B), the process (1) is performed. Further, as in (C) to (F), when P (n−1) and P (n) are unsaturated partial periods and have the same rotation direction, the process of (2) is performed. . In the case of (G) to (I), P (n-1) or P (n) has a saturated partial period or the inversion of the rotation direction is detected, so the process of (3) is performed. .

(1) Current vehicle speed calculation with high-precision acceleration compensation calculation In this case, the post-acceleration compensation period is calculated as follows.
Cycle after acceleration compensation = steady speed term + first correction term + second correction term

The steady speed term is calculated as follows.
Steady speed term = P (n) + P (n-1) + P (n-2) + P (n-3) + P (n-4) + P (n-5)
In this way, the sum of partial periods corresponding to one period of the Hall signal is obtained.

The first correction term is expressed as follows.
First correction term = (6/2 + 1/2) × {P (n) −P (n−6)} / 6
= {P (n) -P (n-6)} × (7/12)
The above “6/2” is an average delay for six partial periods and represents a speed detection delay. “1/2” is an average delay of the PWM output that is output immediately after detection and is held until the output is updated by the next pulse, and represents a control delay. That is, the first correction term represents a delay component proportional to the pulse period.

Further, the second correction term is expressed as follows.
Second correction term = {P (n) −P (n−6)} × {Lf × period count frequency (= constant) / (value of steady speed term)}
The delay time Lf (= frame latency) is, for example, 10 msec. In the second correction term, the value of the steady speed term (P (n) + P (n−1) + P (n) is converted into one period of the Hall signal. -2) + P (n-3) + P (n-4) + P (n-5)).

Specifically, the delay time Lf is determined as follows.
Lf = {Kfl / frame frequency} seconds Kfl = number of equivalent delay frames = average delay 0.5 frame from reflection of the Hall signal to PWM + average delay 0.5 frame by maintaining one PWM frame + calculation delay α frame = 1 + α
0 ≦ Kfl <2
Standard expectation value = 1 + α (α is the delay for calculation, etc.)
Thus, Kfl is set as an adjustable constant.

If the period after acceleration compensation is obtained,
1 / (cycle after acceleration compensation x number of hall cycles per rotation)
The rotational speed is obtained by. Further, the vehicle speed can be obtained by multiplying the circumference of the motor drive wheel.

  In the present embodiment, the difference (change amount) of the partial period from the latest partial period to one period before the Hall signal is calculated as an average acceleration, such as {P (n) -P (n-6)}. However, the acceleration may be calculated by another partial period.

(2) Current vehicle speed calculation with acceleration compensation calculation
In the case of (2), since only P (n) and P (n-1) are normal data, calculation is performed as follows.
Cycle after acceleration compensation = P (n) × 6 + (P (n) −P (n−1)) × 6
= 6 (2P (n) -P (n-1))
In the present embodiment, the change amount of the partial period of (P (n) −P (n−1)) is calculated as the average acceleration.

If the period after acceleration compensation is obtained,
1 / (cycle after acceleration compensation x number of hall cycles per rotation)
The rotational speed is obtained by. Further, the vehicle speed can be obtained by multiplying the circumference of the motor drive wheel.

(3) Calculation of the current vehicle speed In the present embodiment, in the state as shown in (3), the cycle is calculated as follows. Acceleration compensation is not performed. At this time, when P (n) itself is in the saturation period, or when inversion occurs in the period of P (n) itself, the saturation period is used as P (n).
Period = P (n) × 6
If the period is obtained in this way,
1 / (period x number of hole periods per revolution)
The rotational speed is obtained by. Further, the vehicle speed can be obtained by multiplying the circumference of the motor drive wheel.

  As described above, in the present embodiment, the current vehicle speed is calculated based on the time measurement using the edge of the hall signal as described above. It is possible to measure without being affected by.

  Therefore, according to the motor drive control device 102, a period in accordance with the real time at which the drive control reaches the motor is calculated and used without causing a delay of a predetermined time or a fixed period as in the conventional example. Therefore, it is possible to respond quickly to changes in pedal speed. As a result, even when a sudden start operation or a sudden stop operation in which the vehicle speed changes rapidly, the speed calculation result is not significantly different from the current speed when the drive signal actually reaches the motor. Thus, the motor 106 can be controlled without excessive current flowing through the motor 106 as in the prior art, and the circuit is not damaged by the excessive current, or an accident due to a sudden change in operation does not occur.

[Embodiment 2]
In the first embodiment, since the current vehicle speed is calculated based on the hall signal, there is a problem if a malfunction occurs in the hall element that outputs the hall signal. Therefore, as shown in FIG. 11, an error detection unit 360 that performs error detection using the partial cycle output from the partial cycle detection unit 311 is provided. When an error detection output is made, the control unit 300 performs error handling processing. For example, the control may be performed based on the vehicle speed estimated from the output of the pedal rotation sensor. Further, repair may be urged by performing a warning display on the operation panel 105.

As a first method, the error detection unit 360 in this embodiment determines an error using the following determination formula.
Determination value = {(P (n) + P (n-6)) / 2} / {(P (n) + P (n-1) + P (n-2) + P (n-3) + P (n-4) + P (n-5)) / 6}

  That is, the ratio of {(P (n) + P (n-6)) / 2} to the average partial period for one hall signal period is calculated. {(P (n) + P (n-6)) / 2} is the average of the latest partial period and the partial period one hall signal before the latest partial period. However, a weighted average may be used instead of a simple average.

  If this determination value does not fall within the allowable range of 1 ± 0.1, for example, an error is determined.

As a second method for detecting an error with higher accuracy, the following determination formula including acceleration compensation is used.
Determination value = {(7/12) × P (n) + (5/12) × P (n−6)} / [{P (n) + P (n−1) + P (n−2) + P (n -3) + P (n-4) + P (n-5)} / 6]

  The average of 6 partial periods in the denominator of this judgment formula has a detection average delay of 3 partial periods, but the average delay of 2 partial periods of the numerator is (0.5 + 6.5) /2=3.5 partial periods. And a shift of 0.5 period occurs. At constant speed, there is no problem, but during acceleration / deceleration, the time difference is shifted. Therefore, P (n) and P (n) The internal ratio of P (n-6) is changed from 3: 3 to 3.5: 2.5. That is, 7/12 and 5/12 are used as weights.

  Even in the first method and the second method, an error is output when an error continues for a predetermined time such as 1 second.

  In the present embodiment, the Hall signal output from the rotation sensor of the motor 106 is used to determine whether a rotation sensor error has occurred or the rotation sensor has failed. Become.

  Although the embodiment of the present invention has been described above, the present invention is not limited to this. For example, the configuration of the control unit 300 may be implemented by hardware or a combination of a program and a processor. The functional block configuration described above is an example, and may not necessarily match the actual circuit module configuration or program module configuration.

  Further, in the above-described embodiment, the description of the processing based on three Hall signals (Hall-U, Hall-V, Hall-W) whose electric angle is shifted by 120 ° in accordance with the rotation of the motor 106 is performed. However, the same processing as described above may be performed using any type of rotation sensor that outputs two or more arbitrary number of phase detection signals that are out of phase.

  Moreover, although the example of PWM control was shown above, it does not necessarily need to be PWM control. PNM (Pulse Number Modulation), PDM (Pulse Density Modulation), 1-bit amplifier, or other methods may be used.

  The motor drive control device can be applied to an electric vehicle including an electric motorcycle, an electric vehicle, a hybrid vehicle, a train, an elevator, an escalator, and other devices that perform motor drive control in an electric motor drive in addition to an electric assist bicycle. It is.

DESCRIPTION OF SYMBOLS 101 Secondary battery 102 Motor drive control apparatus 103 Torque sensor 104 Brake sensor 105 Operation panel 106 Motor 300 Control part 400 Drive part 410 Drive control circuit 420 Drive circuit 310 Vehicle speed calculation part 320 2nd duty ratio conversion part 330 1st duty ratio conversion Unit 350 target torque calculation unit 311 partial period detection unit 312 current vehicle speed calculation unit 360 error detection unit

Claims (3)

A measurement unit that receives a periodic signal that is received from the rotation sensor of the motor and can be divided into a plurality of phases according to the rotation phase of the motor, and measures time for each of the plurality of phases;
Using the time for each of the plurality of phases measured by the measurement unit, the average time for one phase obtained from the time for one cycle of the signal, and the time of a phase of the plurality of phases And a determination unit that outputs an error of the rotation sensor when a predetermined condition is satisfied by comparing an average or a weighted average with a corresponding phase time one cycle before the signal from the certain phase,
Control device. The determination unit is
The control device according to claim 1, wherein it is determined whether a ratio between the average time and the average or weighted average is within a predetermined range. The control device according to claim 1, wherein the determination unit outputs an error when a state satisfying the predetermined condition continues for a predetermined time.

Images