JPH04251592A - Controller of wheel driving motor for electric vehicle - Google Patents

Abstract

PURPOSE:To output a current command having a voltage pattern corresponding to the operational command for a motor vehicle by predicting the pole position so that a voltage difference, which appears when there is a difference between an actual pole position and a pole position of motor calculated based on a rotational speed, may be eliminated. CONSTITUTION:DC currents, based on a voltage Vdc from a power supply 7, are outputted, as U, V, W phase currents, through each wheel driving motor control circuit 1. Each of U, V, W phase voltages is detected through an induced voltage measuring circuit 10. The induced voltage measuring circuit 10 corrects the pole position of a motor 9 based on these voltages and delivers a detection signal to a high speed operating processor, i.e., a DSP(digital signal processor) 3. Relationship between a position error and voltage is derived for a case where the pole position assumed in the DSP 3 is different from an actual pole position, and then prediction of pole position is repeated so that the position error converges to zero.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electric vehicle, and more particularly to a motor control device for controlling an electric vehicle drive motor without a rotor position sensor for the drive motor.

0002

[Prior Art] Conventionally, when a brushless DC motor is used as a wheel motor (hereinafter simply referred to as a motor) for driving the wheels of an electric vehicle, the absolute position of the magnetic pole, which is the rotor of the motor, is detected, and the Analog control is performed by comparing the current command value that determines what kind of current should flow to a certain position with the current that actually flows. A resolver or absolute encoder is used as a magnetic pole position detector.

FIG. 24 shows a configuration diagram of a motor control section of an electric vehicle using a resolver circuit. As shown in FIG. 24, the motor control section includes a motor drive current waveform control means 23 that controls the current waveform supplied to the motor based on a motor control command from a vehicle control device (not shown), and a motor state necessary for controlling the motor. and means 24 for detecting information. Here, the motor drive current waveform control means 23 is composed of a control circuit 230, a base drive circuit 231, and a bridge circuit 232. Further, the detection means 24
are current sensor 240, resolver 241, resolver circuit 2
42. The control circuit 230 generates a required drive current according to the motor control command signal and a current sensor 240.
A UVW phase pulse width modulation (PWM) signal is output to the base drive circuit 231 based on the deviation from the UV phase current. Here, the motor control command signal includes a current command to generate torque according to the accelerator opening, a rotation direction command according to the forward/reverse signal, a regeneration command according to the amount of brake depression, a driving command to drive the motor, etc. be. Additionally, information required by the vehicle control device is output from the control circuit as state information required for motor control. For example, the information includes the rotational position of the magnetic pole, the temperature of the transistor in the bridge circuit 232, and an operation permission that indicates the completion of startup of the motor control unit.

[0004] The resolver circuit 242 processes a resolver signal representing the operating state of the motor detected by the resolver 241 and outputs a magnetic pole position signal to the control circuit 230. That is, the resolver 241 is excited by flowing current from the resolver circuit 242, detects a magnetic change between it and the magnetic pole of the motor 22, converts it into an electric signal, and generates a resolver signal.

[0005]

[Problems to be Solved by the Invention] When a brushless DC motor is used as a drive motor for an electric vehicle, a position detector is required, but the position detector requires increasing the axial length of the wheel drive shaft and downsizing the motor. Can not do it. In addition, the resolver and absolute encoder are expensive,
Another drawback is that it is easily influenced by environmental factors such as vibration and temperature. Furthermore, when used in a vehicle, the motor control device and the motor body must be placed in separate locations, resulting in long signal lines.
This can lead to negative effects such as becoming less sensitive to noise. In electric vehicles, bridge circuits are driven at high voltage and high frequency, and are a major source of noise. Furthermore, there is a risk that the signal line may be disconnected. Furthermore, since the position detector described above is analog controlled, it also has the disadvantage that sophisticated control cannot be performed.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a wheel drive motor for an electric vehicle using a brushless DC motor that does not use a magnetic pole position detector.

[0007]

[Means for Solving the Problems] The above objects of the present invention are achieved by the following configuration. That is, in a wheel drive motor control device for an electric vehicle that uses a brushless DC motor that outputs DC voltage from a power source as a multiphase current to the drive motor of each wheel, the current output to each wheel drive motor is The rotational speed of the motor is determined based on the rotational speed, and the magnetic pole position is estimated so as to eliminate the voltage deviation that occurs when there is a deviation between the magnetic pole position of the motor calculated based on the rotational speed and the actual magnetic pole position. This is a wheel drive motor control device for an electric vehicle, characterized by comprising a wheel drive motor output control means for outputting a current command value having a voltage pattern according to a driving command of the electric vehicle.

[0008]

According to the present invention, the magnetic pole position is estimated and the speed of the wheel drive motor is adjusted by measuring the induced voltage caused by the rotation of the magnetic pole without using a magnetic pole position detector. That is, the magnetic pole position of the motor at the time of starting is estimated, and a voltage pattern corresponding to the estimated value of the magnetic pole position is output to each phase. Furthermore, when the motor starts rotating, the rate of change in the magnetic pole position of the rotating motor (the rotational speed of the magnetic pole ), and the estimated value of the magnetic pole position is corrected so that the induced voltage caused by the deviation between the estimated value of the magnetic pole position and the actual position becomes zero. The motor is controlled by outputting a current command value corresponding to the deviation between the rotational speed of the magnetic pole obtained as a result and the speed command value. The magnetic pole position can be stored in the memory of the wheel drive motor output control means.

Thus, the present invention provides brushless D for electric vehicles.
As a drive motor using a C motor, the motor can be downsized by sensorless control, and the advantages of a brushless DC motor that is resistant to environmental factors such as vibration and temperature can be utilized. Furthermore, since sensorless control is possible, only the power source is interposed between the wheel drive motor control device and the motor body, so there is no risk of noise, disconnection of signal lines, or failure of the sensor itself.

0010

DESCRIPTION OF THE PREFERRED EMBODIMENTS Examples of the present invention will be described with reference to the drawings.

FIG. 1 shows a control block diagram of the electric vehicle of this embodiment. A DC current based on the voltage Vdc from the motive power source 7 is outputted as a UVW phase current via each wheel drive motor control circuit 1. The motor control circuit 1 is controlled by a motor control command from a vehicle control device (main computer) 2 based on a UVW phase pulse width modulation (PWM) signal outputted via a DSP circuit 3 to be described later. The motor control circuit 1 includes a well-known base drive circuit 5 and a bridge circuit 6, as shown in FIG. Outputs a transistor drive signal that controls the conduction time.

Further, as shown in FIG. 2, the bridge circuit 5 includes a pair of transistors for each phase connected in series to the DC power source 7, and connects the emitter of one transistor of each phase to the collector of the other transistor. Then, the current of each phase is taken out from this connection point, and the three-phase current is supplied to a DC brushless motor 9 for a wheel, for example, with a rating of AC 40V and 120A. The base of each transistor is connected to a base drive circuit 6, a transistor drive current is supplied as a base current, and conduction of the transistor is controlled. By controlling this pair of transistors, the magnitude and width of the phase current can be changed. As shown in Figure 1, U generated in each motor when the ignition is turned off
The voltage of each phase of VW is detected by an induced voltage measuring circuit 10. Based on this voltage, the induced voltage measuring circuit 10 outputs this detection signal to the DSP circuit 3 in order to correct the position of the magnetic pole (not shown) of the motor 9.

In FIG. 1, the DSP circuit 3 outputs a UVW phase PWM signal to the motor control circuit 1 based on the deviation between the U-phase and V-phase currents detected by the current sensor 11 and the required drive current from the vehicle control device 2. Further, the motor driving DC voltage Vdc is output to the DSP circuit 3 for calculation of UVW phase voltages.

FIG. 2 shows a configuration diagram of the motor control section for each wheel, but the difference between the motor control section shown in FIG. 24 and the motor control section is the magnetic pole of the motor status information detection means 24 shown in FIG. Instead of the position detection resolver 241, resolver circuit 242, and control circuit 230, the induced voltage measurement circuit 10 and the DSP circuit 3 are used. In the power supply circuit 13, the direct current supplied from the power (DC) power supply 7 is stabilized to create a constant voltage bridge drive power supply and a control power supply that supplies the operating current of each circuit.

Next, an outline of sensorless control of the brushless DC motor 9 without a position detector using a DSP (digital signal processor) circuit 3, which is a high-speed arithmetic processor especially suited for signal processing, will be described.

This method derives the relationship between the position error and voltage when the magnetic pole position assumed in the DSP circuit 3 is different from the actual magnetic pole position, and thereby estimates the magnetic pole position so that the position error converges to zero. Control is performed by repeating the steps.

At this time, an analytical model of the brushless DC motor 9 shown in FIG. 3 is used to simplify the calculation. Here, the magnetic pole made of a permanent magnet is represented by an excitation winding that is excited with a direct current (constant direct current excitation) that is required when an electromagnet virtually generates the same magnetic field as that generated by a permanent magnet. . In addition, the d-q axes in the figure are orthogonal coordinate axes corresponding to the actual magnetic pole position, and the γ-δ axes are orthogonal coordinate axes to the assumed magnetic pole rotation axis (the d-q axes correspond to the actual magnetic pole position). The coordinate axes are fixed to the magnetic poles.
If the assumed magnetic pole position matches the actual magnetic pole position,
The γ-δ axis overlaps and coincides with the d-q axis. ). Also, d
θ/dt and dθc/dt indicate the rotational speed of the d-q axis and the γ-δ axis, respectively, as given by formula (5),
Δθ indicates a deviation angle between both coordinate axes, that is, a positional deviation. The voltage equation on the three phases of this model is expressed in the form of equation (1) when the phase order etc. are determined as shown in FIG.

[0018]

[Equation 1] Consider projecting Equation (1) onto the γ-δ axis in FIG. As mentioned above, the γ-δ axis is a coordinate axis corresponding to the estimated position angle θc of the DSP circuit 3, and here it is assumed that it has a deviation from the actual position angle θ by the formula (2). .

[0019]

[Equation 2] Furthermore, the three-phase → γ-δ axis transformation matrix can be calculated using the formula (
3) and is given as shown in equation (4).

[0020]

[Math 3]

[Equation 4] From Equation (3) and Equation (4), Equation (1) can be rewritten into Equation (5) by setting the excitation current to if=If=constant. Note that If is a virtual electromagnetic current necessary to generate the same magnetic field as a permanent magnet.

[0021]

[Equation 5] Equation (5) is as follows: Actual applied voltage V3=[Vu Vv V
w] indicates the γ-δ axis component of T, and is a value obtained by coordinate transformation of each phase voltage using Equation (3) and Equation (4). If the actual magnetic pole position θ and the estimated value θc match,
The γ-δ axis is equal to the d-q axis, and the current iγ=id, i
δ=iq. iq is an active current that produces motor torque, and id is a reactive current that does not produce torque.

This sensorless control is performed in accordance with the following steps a to c based on equation (5). a. The estimated rotational speed dθ/dt of the motor 9 at the present moment is obtained by solving Equation (5) by giving the detected values of voltage and current, and the rotational speed dθc of the γ-δ axis is further corrected in the following b. /dt is obtained. b. a
By integrating the internal velocity dθc/dt (velocity calculated by estimation in the DSP circuit), the position angle θc of the γ-δ axis is calculated.
get. If there is a deviation between θc and θ, its influence will appear in the voltage information. Therefore, the deviation between the estimated position and the actual position is determined from the voltage error information, and the position error is converged to zero by adjusting dθc/dt. c. Based on the position angle θc of the γ-δ axis obtained from a and b, the current command Ip
Coordinate transform (inverse transform) of * into three-phase current and determine the current command value for each phase.

A specific application example of the sensorless control described above will be explained next. Current command value Ip corresponding to the amount of accelerator depression from time (n-1) to time n in each phase of the three-phase current output to the bridge circuit 6 shown in FIGS. 1 and 2
*(n-1) (Fig. 4(a)) and the actual output current value at the estimated value θc of the magnetic pole position are compared, and the deviation between the actual output current and the current command value is the set value (dimax/2) If the deviation does not reach the set value, select the 0 voltage vector so that the actual current does not exceed the command value and deviate significantly in the opposite direction by applying voltage, and if the deviation exceeds the set value, select the optimal voltage output to reduce the deviation. Select a pattern. Note that time n and time (n
-1) is as short as 200 μs, for example, so the current command value Ip*(n-1) during this time is set to a constant value. The current sampling during the control period Tc at this time is shown in FIG.
It is carried out at the six times B to F shown in (b). The inventors named this method the improved instantaneous value comparison current control method, and a conceptual diagram thereof is shown in FIG.

Assuming that the current command value is I*sinωt, an output voltage is selected such that the triangular wave-shaped actual current I approaches I*sinωt. The voltage output pattern selected at this time is as shown in FIG. 6, and when the output pattern in this figure is represented by a vector, it becomes as shown in FIG.

Note that in FIG. 4(b), time (n-1
) to Td is a delay time due to control calculations and dead time, and in order to avoid the influence of noise due to pattern changes, sampling can actually be performed only from time B.

In addition to the instantaneous current value comparison method shown in FIG. 5, the triangular wave comparison method described below can also be applied to this embodiment, similar to the current control method using an analog circuit. A conceptual diagram is shown in FIG. In the analog circuit of FIG. 8(a), the deviation between the current command Ip* and the actual current i is subjected to PI control, and then the command voltage e is compared with the triangular wave carrier voltage to determine the output voltage. When the triangular wave comparison method is applied to this example, the command voltage e and the triangular wave carrier voltage are calculated by software, and as shown in FIG. 8(b), when the command voltage has a value larger than the triangular wave carrier voltage. This is a method of turning on the output voltage.

In this embodiment, the motor drive control of the wheel drive motor of the electric vehicle is performed separately in a motor drive initial setting mode and a motor drive control mode after motor drive has started. The former is used for electric vehicle ignitions (
Not shown. ) is a mode in which the output voltage pattern is set according to the latest magnetic pole position θ obtained in advance when the electric vehicle is turned on from off. This is a mode in which the rotational speed dθ/dt and position θ of the magnetic pole are adjusted while outputting a voltage pattern based on a current command value corresponding to the command.

Next, the concept of magnetic pole position fluctuation correction according to the present invention will be explained. The outline is as follows. That is, when the wheel drive motor is rotated by an external force when the ignition of an electric vehicle is turned off, the induced voltage generated in the coil of the motor is measured, the magnetic pole position is estimated from the waveform, and the wheel drive motor is rotated. It is stored in the memory of the output control means. However, if the position changes within π/2 in terms of electrical angle, the motor can be driven even if it does not rotate according to the command, so in that case, the magnetic pole position is not corrected, but is controlled while the ignition is on. Correct. Further, if there is no possibility of the position changing while the ignition is turned off, the magnetic pole position is not estimated and the stored value of the magnetic pole position in the memory is used when the ignition is turned on.

Further, to explain in more detail, the magnetic pole position when the power of the vehicle control device 2 is turned off is stored, for example, in the memory of the DSP circuit 3;
If the wheel rotates due to an external force for some reason, the stored value of the magnetic pole position in the memory will differ from the actual position. Therefore, in order to determine the actual latest position of the magnetic pole, a circuit is used that measures the induced voltage that occurs when the magnetic pole position changes due to an external force. The method of correcting the magnetic pole position of this embodiment is based on the magnetic pole position θ and its rotational speed d that have changed due to external force as shown in FIG.
The magnetic pole position θ is determined by measuring the induced voltage generated by θ/dt.

Next, a method for determining the magnetic pole position θ will be explained. Assume that when the vehicle control device 2 is powered off, the wheels are rotated by an external force or the like, and an induced voltage is generated in the U phase due to a change in the magnetic pole position having a waveform as shown in FIG. 10, for example. For example, since the vehicle speed is 10 km/h and the motor rotation speed is about 300 rpm, an induced voltage of about 10 V is generated. Assuming that the final extreme value of the induced voltage at this time is V0 and the final voltage is VE, it can be assumed that V0 and VE are on the same sinusoidal curve as shown in FIG. Then, since the angle θE at the end of the magnetic pole fluctuation can be calculated from the ratio of the absolute value of the extreme value VE and the final extreme value V0, the end position Δα of the magnetic pole fluctuation can be found.

In FIG. 12(a), θs is the electrical angle when the magnetic pole position fluctuation occurs, and VS is the induced voltage at that time. This magnetic pole position correction is performed for all of the U, V, and W phases. Moreover, as shown in FIG. 12(b), U,
If the extreme value V0' is not obtained for both the V and W phases, the electrical angle of magnetic pole variation is less than π/2, so even if it is treated as if there is no positional variation of the magnetic pole, problems such as reverse rotation of the motor 9 will occur. There's no fear.

Next, FIG. 1 shows the motor control procedure of this embodiment.
This will be explained using the flowcharts shown in FIGS. 3 to 22.

First, the flow of the main routine will be explained with reference to FIG. When the power is turned on, a driving command is read from the vehicle control device 2 (Fig. 1). If there is a driving command, the initial setting of the motor drive (step 3) and motor drive control (step 4) are performed. If there is no driving command, the magnetic pole is set. After correcting the positional fluctuation, the power is turned off (steps 5 and 6).

Flowcharts of the motor drive initial setting mode are shown in FIGS. 14 and 15. First, when the ignition is turned on, the time (n-1
) to time n, a timer is set for the first time B, and then the latest magnetic pole position θck stored in the memory, etc. in the DSP circuit 3 is read. (Step 102). Here, k is an integer of 1, 2, . . . , M, and M is equal to the total number of motors. Therefore, the following flow is performed for all motors 9. Then, data on the current command Ip*k corresponding to the amount of accelerator depression and the wheel rotation direction Rk are read, and the current command signal Ip*k=sgn(Rk)Ip
*k is output (step 104). In addition, sgn(R
k) has a positive value when the wheels are driven forward, and a negative value when the wheels are driven backwards.

Next, the three-phase currents iuk and iv (current=0) at time (n-1), which is the start of motor drive in FIG.
k, iwk are detected, and based on the results and the magnetic pole position θck read in step 102, it is determined which of the voltage patterns shown in FIG. 6 is most appropriate for the magnetic pole position θck. Step 106). As mentioned above, this is called an improved instantaneous current value comparison. Step 1
The optimum voltage pattern is output based on the result of step 06. At this time, the DC voltage Vdc (FIGS. 2 and 3) is detected in consideration of the change over time in the output voltage of the battery 7 of the electric vehicle, and the actual applied voltages Vuk, Vvk, and Vwk are calculated using this battery voltage Vdc as a reference. (Step 109)
). The calculation result of this actual applied voltage is
, D, E, and F times to calculate the two-phase voltage on the γ-δ axis (step 113).

[0036] Next, every time times B to F occur in the current detection cycle (ΔT), interrupt processing is performed to obtain necessary data in the motor drive control mode from time n to time (n+1) (step 110). ). That is, the three-phase currents iuk, ivk,
iwk is detected, and when the Tc time has elapsed (=time n), the process moves to the next motor drive control mode (step 4). The current values measured at times B to F are used for calculations in steps 308 to 314 in FIGS. 16 to 17. Each interrupt process converts the three-phase detection current into two-phase currents iγk, i on the γ-δ axis.
Calculation is performed to convert into δk, and two-phase voltages Vγk, V on the γ-δ axis are calculated from the actual applied voltage calculated in step 109.
Calculate δk. It is assumed that the actual position of the magnetic pole when calculating the two-phase voltage and the two-phase voltage is at θck read from the memory in step 102.

Next, a flowchart of the motor drive control mode from time n to time (n+1) shown in FIGS. 16 to 18 will be shown. Again, first, time (n+1) to (n+2)
Set a timer at time B to detect the current between
Enters motor drive control. A vehicle driving command signal is read from the vehicle control device 2, and if the vehicle driving command is off, magnetic pole position fluctuation correction in step 5 using the induced voltage measuring circuit 10 is performed. When the driving command is on, the current command Ip*k corresponding to the accelerator depression amount and the rotation direction Rk are read, and the current command Ip*k=sgn(Rk)·Ip*k is output. Then, this current command Ip*k and step 1
Compare the three-phase current value detected in step 11 with the current at time F,
Instantaneous current values are compared and an optimal pattern is output (step 307). Then, based on the two-phase currents and voltages at points B, D, and F calculated in steps 112 and 113, the rate of change from points B and F is determined using point D as a reference, and assuming that the magnetic pole is at the θck position. , calculate the estimated speed dθk/dt on the γ-δ axis at point D. Also, the estimated speed dθ at point D
From k/dt, the two-phase currents and two-phase voltages at points B, D, and F, the voltage deviation that occurs when the actual position of the magnetic pole is not equal to θck (
The induced voltage) ΔVγk is calculated (step 309). Next, three-phase currents iuk, ivk, and iwk at time n are detected. Then, this detected value is converted into two phases, and the estimated speed dθk/dt at point E is calculated from this value and the two-phase currents and voltages at points C and E, assuming that the actual position of the magnetic pole is at θck. Also,
From the estimated speed dθk/dt at point E and the two-phase current and voltage at point C and E, calculate the voltage deviation ΔVγk at point E that occurs when the actual position of the magnetic pole is not equal to θck (step 31
3). In this way, dθk/dt and △V at points D and E
From γk, the magnetic pole rotation speed dθck/dt is calculated using the following formula (6).
(step 314). dθck/dt=dθk/dt−α×△Vγk×sgn
{dθk/dt} (6) However, α is a constant greater than or equal to zero, and sgn{X} represents 1 when X>0 and −1 when X<0.

Next, as in steps 108 and 109, the battery DC voltage Vdc and the actual applied voltages Vuk, Vvk, Vw
By calculating k, the two-phase current between the next time n and (n+1),
In order to calculate the voltage, the data in steps 319 to 322 are used.

Then, the magnetic pole position θck is estimated by integrating the magnetic pole velocity dθck/dt obtained in step 314,
Store this value in memory (steps 317, 318
). Next, a flowchart of comparing instantaneous current values in steps 106 and 306 is shown in FIG.

First, the current difference criterion dimax, which will be described later, is
In order to correct the influence of the induced voltage of /2, the current command I
Similar to the p*k induced voltage compensation, resistance drop correction is performed to correct the influence of the dimax/2 resistance drop. Then, assuming that the magnetic pole position is at θck, the current command iγ
k=0 (If the γ-δ axis and the d-q axis match, ideally iγ=id=
Think 0. ), iδk=Ip*k is converted into three-phase current commands iuk*, ivk*, and iwk*. Actual current in each phase (step 105 and step 110
The actual current value detected at time F between time (n-1) detected at and time n is used. ) and the command current. Then, in order to prevent the actual current from exceeding the command current and deviating greatly in the opposite direction by applying a voltage, the phase where the current difference is the maximum is found, and the current difference of that phase is dimax/ If it does not exceed 2, then (
000) or (111) to allow the magnetic pole to rotate by inertia without outputting the voltage (step 40).
8). Furthermore, when dimax/2 is exceeded, an output pattern that works in the direction of reducing the current difference is selected from among the voltage output patterns shown in FIG. 6 (step 407). In addition, here dim
ax/2 represents 1/2 of the maximum value of the amount of current that increases when a voltage is applied during the current control period Tc time.

Next, flowcharts of the magnetic pole position fluctuation correction in step 5 are shown in FIGS. 20 and 21. When the induced voltage measurement circuit 10 detects an induced voltage while the ignition is turned off,
When the power of the DSP circuit 3 is turned on by the induced voltage measuring circuit and no driving command is output to the vehicle, the latest magnetic pole position data θ stored in the memory of the DSP circuit 3, etc.
Load ck. Then, for one motor 9, three-phase voltage data is read from the memory of an induced voltage measuring circuit 10, which will be described later. In this case, when the wheels are rotated by an external force or the like after the ignition is turned off, maximum and minimum voltage values occur in each phase, as shown in FIG. 10. If this extreme value exists, the final extreme value is determined in step 507.
(step 504). If there is one or more phases with extreme values, select that one phase and calculate η=final voltage data (VE)/|final voltage extreme value (VO)| (step 507); As explained with reference to FIG. 11, the variation end position Δα is calculated. and,
Depending on the phase selected in step 506, steps 510-
512, the estimated position of the magnetic pole is calculated. In this way, the estimated magnetic pole position is calculated for all the phases having extreme values, and the average value is stored as the latest magnetic pole position θck in the memory of the DSP circuit 3, for example.

Note that FIG. 22 shows a flowchart in which the memory of the DSP circuit 3 is used as a memory for the voltage measurement value of the induced voltage measuring circuit 10. When an induced voltage is generated in any phase, if the power of the DSP circuit 3 is off, it is turned on and the power measurement values of each phase are written in the memory of the DSP circuit 3. Of course, if there is a memory in the induced voltage measuring circuit 10, it may be used.

In addition, FIG. 23 shows that the three-phase voltage is determined by one of the three phases if the change in the position of the magnetic pole has a width of π/2 or more in terms of electrical angle.
The phase indicates that an extreme value appears in the induced voltage waveform within that width. Therefore, it can be seen that in most cases of magnetic pole position fluctuation, there is an extreme value within the range of the induced voltage change.

As described above, according to this embodiment, the magnetic pole position and speed of the wheel drive motor 9 are estimated by measuring the induced voltage caused by the rotation of the magnetic pole without using a magnetic pole position detector. That is, if the magnetic pole position of the motor 9 at the time of starting is known, the magnetic pole position of the rotating motor 9 can be estimated using the DSP circuit 3 when the ignition of the electric vehicle is turned on. Further, when the ignition of the electric vehicle is turned off, the induced voltage generated in the coil when the motor 9 is rotated by an external force is measured, and the magnetic pole position is estimated from the waveform and stored in the memory of the DSP circuit 3. However, if the position changes within π/2 in electrical angle, the motor can be driven even if it does not rotate as instructed, so in that case, do not correct the position, but correct the magnetic pole position using control while the ignition is on. . Further, if there is no possibility that the position will change while the ignition is turned off, there is no particular need to estimate the position, and it is sufficient to simply store the magnetic pole position in the memory.

The above description can be summarized as follows. a. Estimating the magnetic pole position when the ignition of an electric vehicle is turned on D
This is possible by reading the correct magnetic pole position from the memory of the SP circuit 3. b. Estimating the magnetic pole position when the ignition of an electric vehicle is turned on. This is done by software, regardless of whether the vehicle is running or stopped. Calculate from the value,
This is done by changing the estimated value in a direction where the voltage deviation becomes zero. c. Estimating the magnetic pole position while the ignition of an electric vehicle is turned off The following cases occur when the magnetic pole position changes while the ignition is turned off. For example, when an electric vehicle is moved by an external force such as human strength, when parked on a slope, when the ignition is turned off before pulling the handbrake and the electric vehicle moves, or when the ignition is turned off when starting on a slope. This occurs when the electric vehicle moves because the handbrake is loosened before the handbrake is turned on, or when the ignition is turned off while driving and coasts.

In such a case, the fluctuation of the magnetic pole position while the ignition is turned off is estimated from the induced voltage, and the data of the magnetic pole position is corrected. However, the present invention is not limited to the above embodiment, and there are two methods for sensorless control of the magnetic pole position.

(A) Method for estimating the magnetic pole position when starting the vehicle For example, ■ For some reason when the ignition is turned off,
Even if the magnetic pole position is different from the latest memorized position of the DSP circuit 3, when starting, only one wheel is driven for a minute time and a minute distance with the magnetic pole position unknown, and the induced voltage of the remaining wheels is measured, and the magnetic pole position is determined. A method for correcting position data, ■ A method for estimating the magnetic pole position from the induced voltage waveform at startup and performing sensorless control without correcting the magnetic pole position using the induced voltage measurement circuit 10, ■ A method for controlling the wheels and motor using a clutch. 9 is disconnected, the motor 9 is idled, and the magnetic pole position is determined by the DSP circuit 3.
A method of estimating the position of the magnetic pole using a magnetic pole position sensor, and a method of estimating the position of the magnetic pole only at the time of startup can be applied.

(B) Method of not estimating the magnetic pole position when starting the vehicle For example, a method (2) of mechanically locking the motor 9 while the electric vehicle is stopped to eliminate positional fluctuations can be applied.

[Brief explanation of the drawing]

FIG. 1 is a control block diagram of an embodiment of the present invention.

FIG. 2 is a configuration diagram of a motor control section according to an embodiment of the present invention.

FIG. 3 is a diagram showing an analytical model of the brushless DC motor of the embodiment.

FIG. 4 is a diagram showing the definition of sampling for current detection in the embodiment.

FIG. 5 is a conceptual diagram of the improved instantaneous value comparison current control method of the embodiment.

[Figure 6] FIG. 3 is a diagram showing a voltage output pattern of the embodiment.

7 is a voltage vector diagram of the voltage output pattern shown in FIG. 6. FIG.

FIG. 8 is a conceptual diagram of an improved instantaneous value comparison current control method using a triangular wave according to the embodiment.

FIG. 9 is a conceptual diagram showing the position and speed of magnetic poles in the embodiment.

FIG. 10 is an example of an induced voltage waveform in the embodiment.

FIG. 11 is a diagram showing the principle of magnetic pole position correction using induced voltage in the embodiment.

FIG. 12 is a detailed view of FIG. 11;

FIG. 13 is a main flowchart of motor control in the embodiment.

FIG. 14 is a flowchart of a subroutine for initializing motor drive in the embodiment.

FIG. 15 is a flowchart of a subroutine for initializing motor drive in the embodiment.

FIG. 16 is a flowchart of a motor drive control subroutine in the embodiment.

FIG. 17 is a flowchart of a motor drive control subroutine in the embodiment.

FIG. 18 is a flowchart of a motor drive control subroutine in the embodiment.

FIG. 19 is a flowchart of a subroutine for comparing instantaneous current values in the embodiment.

FIG. 20 is a flowchart of a subroutine for magnetic pole position fluctuation correction in the embodiment.

FIG. 21 is a flowchart of a subroutine for magnetic pole position fluctuation correction in the embodiment.

FIG. 22 is a flowchart of the induced voltage measuring circuit of the embodiment.

FIG. 23 is a diagram showing the induced voltage of each phase in the example.

FIG. 24 is a configuration diagram of a conventional motor control section.

[Explanation of symbols]

1 Motor control circuit 2 Vehicle control device 3 DSP circuit 5 Base drive circuit 6 Bridge circuit 7. Power source 9 Brushless DC motor 10 Induced voltage measurement circuit

Claims (1)

[Claims] [Claim 1] A brushless DC that outputs DC voltage from a power source as a multiphase current to a drive motor for each wheel.
In a wheel drive motor control device for an electric vehicle that uses a motor, the rotation speed of the motor is determined based on the current output to each wheel drive motor, and the magnetic pole position of the motor calculated based on the rotation speed and the actual The wheel drive motor output outputs a current command value with a voltage pattern according to the driving command of the electric vehicle by estimating the magnetic pole position so as to eliminate the voltage deviation that occurs when there is a deviation between the magnetic pole position of the electric vehicle. A motor control device for driving wheels of an electric vehicle, characterized by comprising a control means.