JP2013138559A - Method and device for calibrating angle sensor attached to synchronous motor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to, even under a state where a synchronous motor is mounted on a vehicle, calibrate an angle sensor with a less number of times of execution of the calibration.SOLUTION: A method includes: a step 100 for determining at which quadrant of a true dp coordinate system a d-axis current vector is positioned; a step 104 for setting vectors obtained by calibrating a d-axis current vector at offset angles of two current vectors as a first and second range delimitation vectors, respectively; a step 108 for applying a current of an intermediate vector calibrated using a mean value of the offset angles to a synchronous motor; steps 114 and 116 for determining to which direction the synchronous motor rotates and selecting a range delimitation vector, out of the first and second range delimitation vectors, which is on a rotation direction side; a step for repeating the steps 106 to 116 by setting the range delimitation vector on the rotation direction side and the latest determined intermediate vector as new first and second range delimitation vectors; and a step for setting an intermediate vector at the time when the synchronous motor stops rotating as a true d-axis current vector.

Description

  The present invention relates to a method and apparatus for calibrating an angle sensor attached to a synchronous motor used as a drive source of an electric vehicle or the like.

  In a vector-controlled synchronous motor used in an electric vehicle or the like, the rotor is rotated in synchronization with the rotating magnetic field of the stator. For this reason, it is necessary to accurately detect the angular position of the rotor by the angle sensor.

  However, even if the angle position of the rotor is obtained by the angle sensor, if the mounting position of the angle sensor is deviated from the correct position (hereinafter, the angle of the deviation is referred to as “offset angle”), the detected rotor The angle position becomes inaccurate, and accurate control of the synchronous motor becomes impossible.

  Therefore, conventionally, techniques for correcting the angular position of the rotor detected by the angle sensor by detecting the offset angle as described below (Patent Documents 1 to 7 below) have been proposed.

  According to the technique described in Patent Document 1, a method of aligning resolvers as angle detection means provided in an electric motor is disclosed. The method disconnects the electric motor from the load, commands a d-axis current command and a speed command, operates the electric motor without load in response to these commands, and determines the rotor speed in response to the speed command. , Each step of determining the offset angle of the resolver based on the speed command and the rotor speed when the rotor speed is substantially stabilized.

  Further, the method described in Patent Document 2 sends an electrical signal to the stator so that the rotor is positioned in the vicinity of a predetermined ideal position, measures the position of the rotor by a position sensor, and measures the measured rotor position. And calculating an offset angle associated with a predetermined ideal position from the initial position and calculating an offset angle for a plurality of ideal positions.

  In addition, the brushless motor initial adjustment device described in Patent Document 3 is capable of accurately performing alignment adjustment between the electrical position of the motor when driving the brushless motor and the zero position of the angle detector in a short time. Therefore, a rotation means for rotating the brushless motor from the outside, a rotation angle detection means for detecting the rotation angle of the motor, a torque command section for giving a torque command to a predetermined phase, and rotation by the rotation means are predetermined. An induced voltage detecting means for detecting an induced voltage generated in the phase selected, and a setting unit for setting an angle at which the phase of the induced voltage detected by the torque signal corresponding to the torque command and the induced voltage detecting means coincides as an offset angle; Have

  Further, the motor control device described in Patent Document 4 has a motor control circuit for inputting the rotor angle of the AC motor output from the angle sensor, and the motor control circuit has an input section for the rotor angle reference signal. The calculation means for calculating the angle deviation amount of the angle sensor based on the angle reference signal, the means for storing the angle deviation amount, and the correction for reading the deviation amount from the storage means and correcting the angle based on the deviation amount Means for controlling the motor based on the corrected rotor angle. The synchronous motor is mechanically connected to the engine, and the calculation of the deviation amount is executed when the engine is in operation and the inverter stops energizing the motor.

  A control apparatus for a synchronous motor described in Patent Document 5 includes an inverter that converts direct current to alternating current to supply power to the synchronous motor, current detection means that detects a current flowing from the inverter to an armature coil of the synchronous motor, Voltage detecting means for detecting the voltage applied to the armature coil, position detecting means for detecting the rotational position of the rotor of the synchronous motor, and magnetic pole position for estimating and calculating the magnetic pole position of the rotor based on the detected current and voltage An estimation means; error calculation means for calculating an error between the detected rotational position of the rotor and the magnetic pole position estimated by the magnetic pole position estimation means; and an actual position of the permanent magnet provided on the rotor from the calculated error And a deviation detecting means for detecting a deviation from the detected rotational position of the rotor, and a correcting means for correcting the detected deviation.

  In the control device for the permanent magnet type rotating electrical machine described in Patent Documents 6 and 7, the field direction of the rotor is d-axis and the d-axis in each state where the rotor is rotating at a plurality of different substantially constant rotation speeds. The magnetic pole position correction amount is set to a predetermined temporary setting value while maintaining both the d-axis current command value and the q-axis current command value in the dq vector control in the dq coordinate system with the direction orthogonal to the q-axis as zero. A predetermined arithmetic expression that executes dq vector control processing and uses the d-axis voltage command value and the q-axis voltage command value as variables from the d-axis voltage command value and the q-axis voltage command value obtained when the dq vector control is executed. The error is calculated on the basis of the error and the rotation speed-error characteristic is derived based on the error in each state and the rotation speed at that time. Further, a magnetic pole position correction amount corresponding to the rotational speed of the rotor for correcting the magnetic pole detection position is obtained based on the rotational speed-error characteristic.

  However, any of the above prior arts includes a step of rotating the electric motor or changing the rotational position over a relatively large angle. For this reason, it is difficult to calibrate the angle in a state where the motor is actually incorporated in the vehicle, and these techniques cannot be applied to, for example, a belt drive type or in-wheel type motor.

  The technique described in Patent Literature 1 requires a step of separating the electric motor from the axle (load) (step 405 in FIG. 4) and a step of rotating the electric motor at a constant speed. It cannot be applied to motors.

Further, the technique described in Patent Document 2 requires that the rotor be rotated so as to be positioned at a plurality of ideal positions.
In the technique described in Patent Document 3, an angle at which a torque command to a predetermined phase coincides with an induced voltage detected in a predetermined phase is set as an offset angle. For this reason, the technique requires detecting the induced voltage, which is usually not available with a control unit (inverter) built into the vehicle and can be used to calibrate a motor already built into the vehicle. become unable.

Furthermore, the technique described in Patent Document 4 can calibrate the angle sensor only when the engine is in operation.
Furthermore, since the technique described in Patent Document 5 estimates the magnetic pole position of the rotor based on the detected current and voltage, a voltage sensor is essential.

  Furthermore, the techniques described in Patent Documents 6 and 7 require that the motor be rotated at a constant speed.

US Pat. No. 7596688 (Publication No. 2008/0272731) US Patent Publication No. 2006/0025951 JP 11-299282 A JP 2002-325493 A Japanese Patent Laid-Open No. 9-56199 JP 2010-148270 A JP 2010-148271 A

  In view of the above facts, the present invention provides a method and apparatus for calibrating an angle sensor, which makes it possible to calibrate the angle sensor with a smaller number of executions than in the prior art even when a synchronous motor is incorporated in a vehicle. For the purpose.

  In order to solve the above problems, an aspect of the present invention is a method for calibrating an angle sensor attached to a vector-controlled synchronous motor, wherein the angle sensor is an angular position of a rotor of the synchronous motor. And the rotation direction can be detected, and the method includes a quadrant determination step of determining in which quadrant of the true dq coordinate system the base current vector in the dq coordinate system before calibration is located, and A quadrant definition vector selection step in which the vectors obtained by calibrating the base current vector using the offset angles of the two current vectors defining the quadrant determined in the quadrant determination step are set as first and second range definition vectors, respectively. An intermediate vector determining step for determining an intermediate vector existing between the first and second range defining vectors, and applying an electric current of the intermediate vector to the synchronous motor; A current application step to be used, a rotation direction determination step of determining in which direction the synchronous motor has rotated, and selecting a range definition vector on the rotation direction side of the first and second range definition vectors A range definition vector selecting step, wherein the range definition vector on the rotation direction side and the intermediate vector determined in the immediately preceding intermediate vector determination step are used as new first and second range definition vectors; Regarding the new first and second range definition vectors, the intermediate vector determination step, the current application step, the rotation direction determination step, and the range definition vector selection step are repeated, and obtained in the repetition step and the repetition step. And a calibration step in which the obtained intermediate vector is a true base current vector.

Preferably, the base current vector before calibration is a d-axis current vector. More preferably, it is a positive d-axis current vector.
In the quadrant determination step of the present invention, preferably, four quadrants divided every 90 ° in the dq coordinate system are to be determined, and the currents of the d-axis current vector and the q-axis current vector are respectively supplied to the synchronous motor. The d-axis current is applied based on the rotation direction of the synchronous motor when the current of the d-axis current vector is applied and the rotation direction of the synchronous motor when the current of the q-axis current vector is applied. Each step of determining in which quadrant of the true dq coordinate system the vector is located.

  The quadrant determination step determines that the offset angle of the d-axis current vector is 0 ° or 180 ° when the synchronous motor does not rotate when the d-axis current vector is applied, and the q-axis current It is determined whether the offset angle of the d-axis current vector is 0 ° or 180 ° according to the rotation direction of the synchronous motor when the vector is applied, and the synchronization is performed when the q-axis current vector is applied. When the motor does not rotate, it is determined that the offset angle of the d-axis current vector is 90 ° or 270 °, and the d-axis current vector depends on the rotation direction of the synchronous motor when the d-axis current vector is applied. Is determined whether the offset angle is 90 ° or 270 °.

  Further, the intermediate vector determining step calculates an average value of offset angles of the first and second range defining vectors, and calculates a current vector obtained by calibrating the d-axis current vector with the average value of the offset angles. Determine as. Here, the current vector obtained by calibrating the d-axis current vector with the average value of the offset angle means that the current (before calibration) d-axis current vector is converted into a true d-axis current vector by the average value of the offset angle in the dq coordinate system. This corresponds to the current vector rotated in the approaching direction.

  Further, the repeating step is executed until the synchronous motor does not rotate even when the current of the intermediate vector is applied to the synchronous motor, and the calibration step is performed when the intermediate vector is not rotated. Is used as the offset angle when driving the synchronous motor.

  The synchronous motor can rotate in both directions within a range of clearance larger than the minimum angle detectable by the angle sensor when the synchronous motor is attached to the drive body. In this case, the drive body can be configured as a vehicle wheel, and the synchronous motor can be configured as an in-wheel motor.

  Another aspect of the invention is configured as an apparatus for calibrating an angle sensor by performing the above method. Preferably, the device is a controller that controls the synchronous motor.

  According to the present invention, as long as the angle sensor can detect the angular position and the rotation direction of the rotor of the synchronous motor, the angle sensor can be calibrated. Therefore, it is not necessary to rotate the synchronous motor at a large angle during calibration, and the angle sensor can be calibrated even when the synchronous motor is incorporated in the vehicle, for example, when the vehicle cannot be freely rotated in a stationary state like an in-wheel motor. It is. Therefore, as in the prior art, by connecting the motor to the test bench means, transferring the determined offset angle data to the vehicle controller, disconnecting the motor from the test bench means, attaching the motor to the vehicle body, etc. A series of steps can be omitted, and the labor and time required for calibration can be greatly reduced.

  In addition, according to the present invention, the range in which the true d-axis current vector exists (the range defined by the first and second range defining vectors) is sequentially narrowed by calculating the intermediate vector. It is possible to significantly reduce the number of required operations and shorten the offset angle calibration time.

FIG. 1 is a schematic diagram of a motor control system to which a method for calibrating an angle sensor according to an embodiment of the present invention is applied. FIG. 2 is a flowchart showing a process flow of a method for calibrating an angle sensor according to an embodiment of the present invention. FIG. 3 is a flowchart showing a detailed processing flow of the initial determination of FIG. FIG. 4 is a flowchart showing a flow of processing for determining the rotation direction. FIG. 5 is a diagram for explaining in which quadrant of the true dq coordinate system the d-axis current vector id is in the initial determination of FIG. 2, and (a) is a case where id is a positive value and a zero value. Indicates the possible quadrant in each case of negative values, and (b) is obtained from the motor rotation directions s1 and s2 when the d-axis and q-axis current vectors id and iq are applied, respectively. The quadrant of id is shown, and (c) is a table showing the relationship between s1 and s2 and offset angles ofs, ofs1 and ofs2. FIG. 6 is a diagram for explaining a repetitive sequence of the angle sensor calibration method of FIG. FIG. 7 is a diagram for explaining the relationship between the dq coordinate system before calibration and the true dq coordinate system and the offset angle. FIG. 8 is a diagram showing an outline of the in-wheel motor. FIG. 9 is a diagram showing an example of attaching a synchronous motor to a vehicle to which the angle sensor calibration method according to the embodiment of the present invention is suitably applied.

FIG. 1 shows a motor control system 1 to which a method for calibrating an angle sensor according to an embodiment of the present invention is applied.
The motor control system 1 includes a controller 2, a vector-controlled synchronous motor 3, an angle sensor 4 that detects an angular position γ * of a rotor (not shown) of the synchronous motor 3, and commands from the controller 2. Based on the inverter 5 that converts the DC power into AC power based on this and supplies it to the synchronous motor 3, and the angle position γ * detected by the angle sensor 5, the offset of the angle sensor 4 obtained by the angle calibration method according to this embodiment. calibrated at an angle gamma ₀ true angular position γ (= γ * -γ ₀₎ and the offset angle calibration device 6 for outputting a voltage in the dq coordinate system outputted from the controller 2 (current) command vector (uq (iq), ud (id)) is converted into a two-phase command in the rotating coordinate system of the angular position γ, and the converted two-phase command is converted into a three-phase command (u1, u2, u3). 2/3 output to the inverter 5 The converter 8, the current sensor 11 that detects a three-phase AC command from the inverter 5 to the synchronous motor 3, and the three-phase detection values (i1 *, i2 *, i3 *) detected by the current sensor 11 into two phases A 3/2 converter 9 for conversion and a converter 10 for converting the converted two-phase detection value into a current vector (id *, iq *) in the dq coordinate system are provided. Based on the motor parameter and the detected current vector (id *, iq *), the controller 2 uses a voltage (current) command vector (uq (iq), ud) for achieving the target torque commanded by the host controller. (id)).

  Examples of the synchronous motor 3 include a permanent magnet synchronous motor, an excitation drive synchronous motor, and a stepping motor, but the present invention is not limited to these examples. Furthermore, the synchronous motor 3 can be configured as an in-wheel motor as shown in FIG. 8, for example. As shown in the figure, the in-wheel motor is configured by incorporating the synchronous motor 3 including the rotor 20 and the stator 21 and the angle sensor 4 into a tire 22 attached to the vehicle body 24. (23 is a tire tube). Even when the vehicle cannot be freely rotated when the vehicle is stationary like an in-wheel motor, the synchronous motor 3 can operate in both directions (clockwise direction and counterclockwise direction) within a clearance larger than the minimum angle detectable by the angle sensor 5 at least. Direction). This clearance can be realized by play or the like existing in the motor power transmission path.

  Note that the angle sensor 4 can detect not only the angular position γ * of the rotor (not shown) of the synchronous motor 3 but also the rotational direction of the synchronous motor 3. Examples of such an angle sensor include an absolute position detection sensor such as an incremental position sensor, a resolver, and a rotary encoder, or a Hall element. However, the angle sensor of the present invention is not limited to these.

  FIG. 9 shows an example of attachment of the synchronous motor 3 to the vehicle. The synchronous motor 3 has a vehicle body as a single motor mounted on a front wheel axle having a differential gear D, two in-wheel motors incorporated in the front wheels, and four in-wheel motors incorporated in the front and rear wheels. Can be attached to.

  Next, a method for calibrating the angle sensor of the present invention in the motor control system 1 of FIG. 1 will be described using the flowcharts of FIGS. 2 to 4 can be executed by the controller 2 or a controller higher than the controller 2.

  FIG. 2 shows the steps of the method for calibrating the angle sensor. As shown in FIG. 2, as an initial determination, it is determined which of the four quadrants of the true dq coordinate system the current d-axis current vector is in (step 100). After it is determined which quadrant the d-axis current vector is in, the process of sequentially narrowing the range in which the d-axis current vector is estimated to exist is continued (steps 102 to 120).

First, detailed processing of this initial determination will be described with reference to FIG.
As shown in FIG. 3, first, parameter initialization is executed (step 200). These parameters include an offset angle ofs, an offset angle (ofs1, ofs2) of the first and second range defining vectors, and an offset determination flag F indicating whether the offset angle has been determined. Each is set to 0.

  Next, the current id of the d-axis current vector is applied to the synchronous motor 3 (step 202). Here, “applying a vector current to the synchronous motor 3” means that in FIG. 1, a command of current id is issued from the controller 2, and the command is converted to the converter 7, 2/3 converter 8, and the inverter 5. This is equivalent to rotating the synchronous motor 3 by this conversion command.

  If the angle sensor is correctly mounted and the d-axis current vector id matches the true d-axis current vector id ′ (ie, the offset angle = 0 °), the motor even if the current id is applied to the synchronous motor 3 Does not rotate. However, as shown in FIG. 7, when the current id deviates from the current id ′ by a certain offset angle, by applying the current id, not only the current component in the d ′ axis direction but also the q ′ axis direction A current component is generated. When this current component in the q′-axis direction is generated, the motor rotates.

  Therefore, the rotation direction S1 of the synchronous motor 3 is determined using the angle sensor 4 (step 204). When the synchronous motor 3 rotates counterclockwise in the dq coordinate system, S1 = 1 is assigned. When the synchronous motor 3 rotates clockwise, S1 = −1 is assigned. When the synchronous motor 3 does not rotate, S1 = 0 is assigned. When S1 = 0 where the motor does not rotate, the current id matches the current id ′ (offset angle = 0 °), or the current id matches the current −id ′ (offset angle = 180). °), but at this stage it is not possible to determine which one is still.

  Next, the current iq of the q-axis current vector orthogonal to the current id is applied to the synchronous motor 3 (step 202). When the current iq of the q-axis current vector matches the true q-axis current vector iq ′ (that is, the offset angle is 0 °), when the current iq is applied to the synchronous motor 3, the motor rotates clockwise. . However, as shown in FIG. 7, when the current id is deviated from the current id ′ by a certain offset angle, by applying the current iq, depending on the offset angle, it rotates counterclockwise in the opposite direction. Or when the offset angle is 90 ° or 270 °. Therefore, the rotational direction S2 of the synchronous motor 3 is determined using the angle sensor 4 (step 208). When the synchronous motor 3 rotates counterclockwise in the dq coordinate system, S2 = 1 is assigned. When the synchronous motor 3 rotates clockwise, S2 = −1 is assigned. When the synchronous motor 3 does not rotate, S2 = 0 is assigned. When S2 = 0 where the motor does not rotate, the current iq matches the current id ′ (offset angle = 270 °), or the current iq matches the current −id ′ (offset angle = 90). In order to determine this, the rotation direction S1 is used as will be described later.

Next, it is determined whether S1 = 0 or S2 = 0 (step 210).
When neither S1 nor S2 is 0 (No at Step 210), the routine proceeds to Step 212. In step 212, the quadrant in which the current id is estimated to exist is determined according to the combination of the positive and negative signs of S1 and S2. This determination method will be described with reference to FIG.

  As shown in FIG. 5A, when S1> 0, the current id exists in quadrant Q1 or Q2 out of the four quadrants Q1, Q2, Q3, and Q4 of the true dq coordinate system, and S1 <0. In FIG. 5A, it exists in the quadrant Q3 or Q4, and when S1 = 0, it is presumed to exist on the + d axis or the −d axis (in FIG. 5A, only one case is displayed). Here, when information on the rotation direction S2 when the current iq is applied is added, it can be determined which of the two candidates exists. In the case of S1> 0, it can be determined whether the current id exists in Q1 or Q2. For example, as shown in FIG. 5B, when S1> 0 and S2 <0, it is estimated that the current id exists in the quadrant Q2. In other words, the current id exists between two range defining vectors iq 'and -id' that define the quadrant Q2. That is, if the offset angles of these two range defining vectors are ofs1 and ofs2, the quadrant Q2 can be expressed by ofs1 = 90 ° and ofs2 = 180 °, and the offset angle of the current id is within the range ofs1 to ofs2. It is in. In FIG. 5C, for each of all combinations of S1 and S2, whether the current id is in quadrant (represented by offset angles ofs1 and ofs2), or + d axis or −d axis ( Data of offset angle ofs = 0 ° or 180 °) is shown. In FIG. 5C, when S1 = S2 = 0, data is not described because it is a possible combination.

  In this way, in step 212, it is determined which of the quadrants Q1, Q2, Q3, and Q4 of the true dq coordinate system the current id is in, and this routine is completed to return to the main routine of FIG. . At this time, the offset determination flag F indicating that the offset angle has been determined remains 0.

  On the other hand, if either S1 or S2 is 0 (Yes determination at step 210), the process proceeds to step 214. In step 214, as shown in the table of FIG. 5C, when S1 = 0, the offset angle = 0 ° or 180 ° is determined according to the sign of S2. When S2 = 0, the offset angle is determined to be 90 ° or 270 ° depending on the sign of S1. Since the offset angle ofs is determined in step 214 as described above, an offset determination flag indicating that the offset angle has been determined is set to 1 (step 216), and this routine is completed to return to the main routine of FIG.

Reference is again made to FIG. After the initial determination detailed in FIG. 3 in step 100 is completed, it is determined whether or not the offset determination flag F = 1 (step 102). When the offset determination flag = 1 (affirmative determination in step 102), since the offset angle ofs has already been determined in step 214 of FIG. 3, the offset angle γ ₀ for angle calibration of the offset angle calibration device 6 is set to ofs. The setting is made (step 120) and the main routine is terminated.

  When the offset determination flag = 0 (negative determination in step 102), the vector offset angles ofs1 and ofs2 defining the quadrant determined in step 212 of FIG. 3 are delivered (step 104).

Next, an average value of the offset angles ofs1 and ofs2 is calculated according to the following equation (step 106).
ofs = (ofs1 + ofs2) / 2
Next, an intermediate vector is generated by calibrating the current id of the current d-axis current vector using the average value ofs of the offset angle calculated in step 106, and the current of the intermediate vector is applied to the synchronous motor 3 ( Step 108). The calibration of the current id by the average value ofs is executed by setting γ ₀ = ofs with the offset angle calibration device 6.

  Next, the rotational direction S of the synchronous motor 3 to which the intermediate vector current is applied is determined (step 110). The sign of S is determined (step 112). If S <0, the average offset ofs calculated in step 106 is substituted for ofs1, and the process returns to step 106. In step 106, the average offset angle ofs is recalculated from the newly calculated ofs1 and ofs2 used in the immediately preceding step 106, and steps 108 to 112 are executed in the same manner as described above. On the other hand, if it is determined in step 112 that S> 0, the average offset ofs calculated in step 106 is substituted for ofs2, and the process returns to step 106. In step 106, the average offset angle ofs is recalculated from ofs1 used in the immediately preceding step 106 and newly calculated ofs2, and steps 108 to 112 are executed in the same manner as described above.

When steps 106 to 112 are repeated, the intermediate vector of step 108 approaches the current id ′, and eventually, in effect, coincides. In this case, even if the intermediate vector current is applied, the synchronous motor does not rotate (S = 0), and ats at this time is regarded as an offset angle formed by the current current vector id with respect to the true current vector id ′. Can do. Thus, the process proceeds from step 112 to step 118, and the offset determination flag F is set to 1 (step 118). Then, the offset angle γ ₀ for angle calibration of the offset angle calibration device 6 is set to ofs (step 120), and this main routine is terminated.

Here, the flow of processing for determining the magnitude of the current applied to the synchronous motor in step 110 of FIG. 2 and steps 204 and 208 of FIG. 3 will be described with reference to FIG.
First, parameters are initialized (step 300). That is, the current id = 0 and the rotation direction S = 0. Next, a predetermined value α (α> 0) is added to the current id (step 302), and the calculated current id is applied to the synchronous motor (step 304). Next, the rotation direction S is determined. S = + 1 for left rotation, that is, counterclockwise, S = −1 for right rotation, that is, clockwise, S = 0 when not rotating (step 306). It is determined whether S = 0 and id <= imax is satisfied (step 308). Here, imax is a value predetermined as the maximum value of the current id. If left and right rotation is observed in step 306, or if the current id has reached the maximum value imax, a negative determination is made in step 308, this routine is completed, and the process returns to the main routine. When the condition of step 308 is satisfied, the process returns to step 302, and α is added again to the current id. Therefore, the current value is larger by α than the initial current id. Steps 304 to 308 are repeated for the new current id until the left and right rotational directions are observed or the current id reaches the maximum value imax. By executing the process of FIG. 4, even if the offset angle can rotate the motor originally, the magnitude of the current id is so small that it can be erroneously determined as “not rotating” without rotating due to the influence of friction or the like. The sex can be reduced.

The repetition sequence of steps 106 to 112 will be described in more detail with reference to FIG.
FIG. 6 shows an example in which it is determined that the current id0 of the current d-axis current vector exists in the second quadrant Q2 of the true dq coordinate system (id′-iq ′). Here, the offset angle ofs formed by the current id0 with respect to the current id ′ is unknown, and since id0 exists in the quadrant Q2, ofs1 = 90 ° and ofs2 = 180 ° are set.

  From the information that the current id0 exists in the quadrant Q2, as shown in FIG. 6, the unknown current id ′ is a first range defining vector idr1 (from id0 rotated 90 ° clockwise by calibrating the current id0 ofs1 = 90 °). And a second range defining vector idr2 (current vector rotated clockwise from id0 by 180 °) calibrated ofs2 = 180 °. Therefore, a vector that is intermediate between the first and second range defining vectors (hereinafter referred to as “intermediate vector”) is obtained as a candidate vector for the current id ′.

  In this embodiment, first, as an intermediate vector, a vector obtained by calibrating the current id0 by the average value ofs = (ofs1 + ofs2) / 2 of the offset angle calculated in step 106 in FIG. 2 (displayed as “id1” in FIG. 6). Is used. As shown in the figure, since the current id1 does not yet coincide with id ', when the current id1 is applied (step 108 in FIG. 2), the synchronous motor rotates. Since the current id1 generates a positive iq 'component, the current id1 has a clockwise rotation direction in FIG. 6 (S = -1). From this rotational direction, the current id1 is estimated to have a positive offset angle counterclockwise relative to id 'and closer to the second range defining vector idr2 than the first range defining vector idr1. Therefore, the current id1 is set as a new first range defining vector (corresponding to step 114 in FIG. 2), and the second range defining vector is used as the same idr2 to obtain the intermediate vector id2 (step 106 in FIG. 2). Equivalent to the recalculation of the offset angle). Here, the new intermediate vector id2 is applied to the synchronous motor. As shown in the figure, the current id2 is closer to id 'than id1, but is not yet coincident with id', so when the current id2 is applied, the synchronous motor rotates clockwise (S = -1). Similarly, the current id2 is set as a new first range defining vector (corresponding to step 114 in FIG. 2), and the second range defining vector is used as the same idr2, and the intermediate vector id3 is obtained.

  When the intermediate vector id3 is applied to the synchronous motor, the current id3 is closer to id 'than id2, but produces a negative iq' component, thus causing counterclockwise rotation (S = + 1). From this rotation direction, it is assumed that the current id3 has a negative offset angle with respect to id ', and id' is closer to the first range defining vector id2 than to the second range defining vector idr2. Therefore, the same id2 is used as the first range defining vector, the current id3 is set as a new second range defining vector (corresponding to step 116 in FIG. 2), and an intermediate vector id4 is obtained.

When the intermediate vector id4 is applied to the synchronous motor, the synchronous motor does not rotate because id4 substantially coincides with id ′. Therefore, ofs calculated at this time can be regarded as the offset angle γ ₀ of the current id ₀ (step 120 in FIG. 2).

  As each step of the repetitive sequence is executed as described above, the difference between the offset angles ofs1 and ofs2 of the first and second range defining vectors decreases, and the difference between the target offset angle and ofs1 (ofs2) decreases. I will do it. When the difference becomes so small that the intermediate vector does not rotate the synchronous motor, the offset angle of the intermediate vector can be set as the target offset angle.

  A method of repeatedly applying an electric current to the motor while changing the offset angle ofs from 0 ° to 360 ° for each predetermined step angle, and finally setting the ofs when the motor stops rotating as a target offset angle Is also possible. In such a method, when the number of executions is n times (maximum n = 360 when the step angle is 1 °), in the embodiment of the present invention, the number of executions of the repetitive sequence = log n, which greatly increases the number of executions. The offset angle calibration time can be shortened by reducing the time.

  The major feature of the embodiment of the present invention is that the angle sensor can be calibrated even when the synchronous motor 3 is incorporated in the vehicle, for example, when the vehicle cannot be freely rotated as in the in-wheel motor of FIG. That's what it means. In the prior art, since it is necessary to rotate the motor at a large angle, if there is a large restriction on free rotation with the motor attached to the vehicle body, such as an in-wheel motor, the angle before attaching the motor to the vehicle body It is necessary to calibrate the sensor. For this reason, in the prior art, a series of tests by experts such as connecting the motor to the test bench means, transferring the determined offset angle data to the vehicle controller, disconnecting the motor from the test bench means, and attaching the motor to the vehicle body. This procedure was necessary, and a great amount of calibration time was spent. In the present invention, these procedures are not required, and the angle sensor can be automatically calibrated by the controller while the motor is attached to the vehicle body.

The above is the embodiment of the present invention, but the present invention is not limited to the above-described example, and can be arbitrarily modified within the scope of the present invention.
For example, the motor control system in FIG. 1 is only an example, and can be applied to other control systems that control a synchronous motor including an angle sensor.

Further, although the positive currents id and iq are applied in steps 202 and 206 of FIG. 3, a negative current -id or -iq may be applied. Further, the order of steps 202 and 206 may be reversed.
Further, the method of calculating the offset of the intermediate vector shown in step 106 of FIG. 2 is not limited to this, and any method can be used as long as it is a method for specifying a vector existing between the first and second range defining vectors. Is possible.

  Furthermore, the synchronous motor of the present invention is not limited to the motor mounting example of FIG. 9, and the angle sensor calibration method of the present invention can be applied to a configuration in which the motor is connected to the axle via a clutch. . Further, the calibration method of the present invention can be applied even when the motor is detached from the vehicle.

DESCRIPTION OF SYMBOLS 1 Motor control system 2 Controller 3 Synchronous motor 4 Angle sensor 5 Inverter 6 Offset angle calibration apparatus 7 Converter 8 2/3 converter 9 3/2 converter 10 Converter 11 Current sensor 20 Rotor 21 Stator 22 Tire 23 Tire Tube 24 Vehicle body

Claims (10)

A method for calibrating an angle sensor attached to a vector controlled synchronous motor comprising:
The angle sensor is capable of detecting an angular position and a rotation direction of a rotor of the synchronous motor,
The method
A quadrant determination step for determining in which quadrant of the true dq coordinate system the base current vector in the dq coordinate system before calibration is located;
A quadrant definition vector selection step, wherein vectors obtained by calibrating the base current vector using offset angles of two current vectors defining the quadrant determined in the quadrant determination step are set as first and second range definition vectors, respectively. ,
An intermediate vector determining step for determining an intermediate vector existing between the first and second range defining vectors;
Applying a current of the intermediate vector to the synchronous motor; and
A rotation direction determination step of determining in which direction the synchronous motor has rotated, and selecting a range definition vector on the rotation direction side of the first and second range definition vectors;
A range defining vector selecting step, wherein the range defining vector on the side of the rotation direction and the intermediate vector determined in the immediately preceding intermediate vector determining step are used as new first and second range defining vectors;
Repeating the intermediate vector determination step, the current application step, the rotation direction determination step, and the range definition vector selection step with respect to the new first and second range definition vectors;
A calibration step in which the intermediate vector obtained in the repetition step is a true base current vector;
A method comprising: The method of claim 1, wherein the base current vector prior to calibration is a d-axis current vector. The quadrant determination step includes
The four quadrants divided every 90 ° in the dq coordinate system are to be determined,
Applying the currents of the d-axis current vector and the q-axis current vector to the synchronous motor,
Based on the rotation direction of the synchronous motor when the current of the d-axis current vector is applied and the rotation direction of the synchronous motor when the current of the q-axis current vector is applied, the d-axis current vector is The method according to claim 2, comprising each step of determining which quadrant of the true dq coordinate system is located. The quadrant determination step includes
If the synchronous motor does not rotate when the d-axis current vector is applied, it is determined that the offset angle of the d-axis current vector is 0 ° or 180 °, and the synchronization when the q-axis current vector is applied. Determining whether the offset angle of the d-axis current vector is 0 ° or 180 ° according to the rotation direction of the motor;
If the synchronous motor does not rotate when the q-axis current vector is applied, it is determined that the offset angle of the d-axis current vector is 90 ° or 270 °, and the synchronization when the d-axis current vector is applied. The method according to claim 3, wherein the d-axis current vector offset angle is determined to be 90 ° or 270 ° depending on a rotation direction of the motor. The intermediate vector determination step includes
Calculating an average value of the offset angles of the first and second range defining vectors;
The method according to any one of claims 2 to 4, wherein a current vector obtained by calibrating the d-axis current vector with an average value of the offset angle is determined as the intermediate vector. The repeating step is executed until the synchronous motor does not rotate even when the current of the intermediate vector is applied to the synchronous motor.
The said calibration process employ | adopts the offset angle of the said intermediate vector when the said synchronous motor stops rotating as an offset angle at the time of drive-controlling the said synchronous motor. the method of. 7. The synchronous motor according to claim 1, wherein the synchronous motor is rotatable in both directions within a clearance range larger than a minimum angle detectable by the angle sensor when the synchronous motor is attached to a driving body. 2. The method according to item 1. The method according to claim 7, wherein the driver is a vehicle wheel and the synchronous motor is configured as an in-wheel motor. Apparatus for calibrating the angle sensor by carrying out the method according to any one of the preceding claims. The apparatus according to claim 9, wherein the apparatus is a controller that controls the synchronous motor.

Images