JP2013251972A - Re-adhesion control method and motor control device - Google Patents

Abstract

A method for solving a problem in calculating a temporary torque return value is proposed.
A motor control device drives a driving wheel by a step return control that lowers a torque command value when it detects the occurrence of idling of a driving wheel, returns the torque command value to a temporary return value, and then returns to the original torque command value. The motor is controlled to re-adhere the wheel. The temporary return value at this time is determined as follows. That is, first, the load torque estimating unit 71 and the load torque holding unit 72 estimate the load torque at the time of idling detection using the acceleration of the moving wheel and the torque command value. Next, the correction unit 80 corrects the estimated load torque based on the idling sliding speed of the driving wheel to determine a temporary return value.
[Selection] Figure 5

Description

  The present invention relates to a re-adhesion control method for re-adhering a moving wheel by controlling an electric motor that drives the moving wheel when occurrence of idling of the moving wheel is detected.

  An electric vehicle, an electric vehicle, and the like are known as a vehicle that travels by driving a driving wheel with an electric motor. An electric vehicle is accelerated and decelerated by a tangential force (also called adhesive force) between wheels and rails. If the driving force generated by the generated torque of the electric motor is less than the adhesive force acting on the wheels and rails, the adhesive running is performed. If the driving force exceeds the adhesive force, the idle running or sliding (hereinafter referred to as “idling running”) is performed. Occurs. When the occurrence of idling is detected, control for reducing the generated torque of the electric motor to return to adhesion running, that is, re-adhesion control is performed (see, for example, Patent Document 1).

The re-adhesion control will be specifically described with reference to FIG. FIG. 1 shows a schematic example of a series of signal waveforms from the state of constant acceleration in which no idling has occurred to the occurrence of idling and readhesion control. With the horizontal axis as time t, in order from the top, a graph showing the axial speed V and the reference speed Vm related to the rotation of the moving shaft (wheel) to be controlled, a graph showing the acceleration A related to the rotation of the control symmetrical axis, and the motor torque It is a graph which shows torque command value (tau) _e ^* . In a state where no idling occurs, the shaft speed V substantially matches the reference speed Vm, and the torque command value τ _e ^* is kept substantially constant. In this state, the torque command value τ _e ^* is the torque pattern value τ _n ^* (original command value) calculated based on the vehicle speed, the notch command, or the like.

  When idling occurs, the shaft speed V starts to increase, and the idling gliding speed (speed difference) ΔV that is the difference from the reference speed Vm increases. Then, when the idle running speed ΔV reaches a predetermined idle running detection threshold Vs at time t1, the occurrence of idle running is detected.

When the occurrence of idling is detected, the re-adhesion control is activated and the torque command value τ _e ^* is lowered. Then, the increase in the acceleration A is gradually suppressed and starts to decrease. During this time, the axial speed V continues to increase, but at time t2 when the acceleration A becomes zero, the increase in the axial speed V also becomes zero. The fact that this acceleration A has become zero is detected as the start of recovery from idle running to the original adhesive running (recovery detection). Although the acceleration A for recovery detection has been described as zero, it has been set to zero for the sake of simplicity, and a predetermined recovery detection threshold (for example, “+1” or “−1” instead of zero). When the following is reached, it may be detected as recovery start.

When the recovery is detected, the reduction of the torque command value τ _e ^* is stopped and held. Then, the decrease in acceleration A, which has been negative, is gradually suppressed, and eventually increases. Further, the axial speed V, which has a large deviation from the reference speed Vm, starts to decrease. When the idling sliding speed ΔV becomes equal to or lower than a predetermined re-adhesion detection threshold Vr, it is detected that re-adhesion occurs (re-adhesion detection), and control for return operation is started. That is, control for gradually returning the torque command value τ _e ^* to the torque pattern value τ _n ^* is started.

In the torque return control, first, a step return control is performed in which the torque temporary return value τ _l ^* is gradually controlled to be the target torque value and then gradually returned to the torque pattern value τ _n ^* . The torque temporary return value τ _l ^* is calculated using the torque pattern value τ _n ^* and the acceleration A. Until the predetermined time elapses until the torque command value τ _e ^* becomes the torque temporary return value τ _l ^* , if the idling is not detected again, it is determined that the temporary return is complete (time t4), and the torque pattern is reached. The torque command value τ _e ^* is gradually returned toward the value τ _n ^* . Then, at time t5 when the torque pattern value τ _n ^* is restored, the re-adhesion control is completed. After the re-adhesion control is completed, the torque pattern value τ _n ^* becomes the torque command value τ _e ^* .

  Note that the monitoring target for the idling / sliding detection and the re-adhesion detection is the axial speed V (and hence the idling / sliding speed ΔV), but the acceleration A may be used in addition to the monitoring target. In addition, although the acceleration A is the monitoring target for recovery detection, the axial speed V (and hence the idling speed ΔV) is also used in addition to the monitoring object, so that the acceleration A becomes zero, or the idling speed ΔV is detected as idling. A method of detecting that the recovery threshold value is below a predetermined threshold value that is equal to or lower than the threshold value Vs is considered as the start of recovery.

JP 2002-44804 A

As a result of investigating each detection value and each control value under the re-adhesion control, it was found that there may be a problem in calculating the torque temporary return value τ _l ^* .

This is a problem related to the calculation of the load torque τ ₁ .
The torque temporary return value τ _l ^* is generally calculated using the torque pattern value τ _n ^* and the acceleration A when the occurrence of idling is detected. As an example, for example, the following equation (1 A method of calculating based on the load torque τ ₁ estimated using
τ _l = τ _n ^* −m × A (1)
Here, m is an equivalent inertial mass, and is a predetermined value determined based on specifications of the electric vehicle such as a wheel radius and a gear ratio. A is the rotational angular acceleration of the moving wheel to be controlled.
The estimated load torque τ _l may be used as the temporary torque return value τ _l ^* as it is, or multiplied by a coefficient k _{D of} 1 or less such as 0.8 or 0.9 as shown in the following equation (2). The temporary return value τ _l ^* may be determined.
τ _l ^* = τ _l × k _D (2)

Originally, the torque temporary return value τ _l ^* should be determined based on the load torque τ _l when the idling actually occurs. This is because the load torque τ ₁ at the time of idling is less likely to cause idling while the torque is as high as possible.

However, when the occurrence of idling is detected, the acceleration A has a certain value (see acceleration A at time t1 in FIG. 1). Therefore, the load torque τ _l estimated by the equation (1) includes an error that the load torque τ _l is estimated as a value smaller than the load torque τ _l when the idling is actually generated. As a result, it can be said that the conventional torque temporary return value τ _l ^* was calculated as a smaller value than the original value (an error was included).

  The present invention has been made for the purpose of proposing a method for solving the above-described problem in the calculation of the temporary torque return value.

The first form for solving the above problems is:
When the occurrence of idling of the driving wheel is detected, the torque command value is lowered, and after returning to the temporary return value, the motor is driven by controlling the motor that drives the driving wheel by the step return control to return to the original torque command value. A re-adhesion control method for re-adhesion,
A load torque estimation step (for example, processing by the load torque estimation unit 71 and the load torque holding unit 72 in FIG. 5) for estimating a load torque at the time of idling detection using the acceleration relating to the rotation of the driving wheel and the torque command value;
A temporary return value determining step (for example, processing by the correction unit 80 in FIG. 5) for correcting the estimated load torque using the idle running speed of the driving wheel or the idle running speed equivalent value to determine the temporary return value; ,
Is a re-adhesion control method.

Further, as another embodiment, an electric motor that drives the driving wheel by step return control that lowers the torque command value when the occurrence of idling of the driving wheel is detected, returns the torque command value to the temporary return value, and then returns to the original torque command value. Is an electric motor control device (for example, electric motor control device 50 in FIG. 4) that re-adheres the moving wheel,
Load torque estimating means (for example, load torque estimating unit 71 and load torque holding unit 72 in FIG. 5) for estimating the load torque at the time of idling detection using the acceleration relating to the rotation of the driving wheel and the torque command value;
Temporary return value determination means (for example, the correction unit 80 in FIG. 5) that determines the temporary return value by correcting the estimated load torque using the idle running speed of the driving wheel or the idle running speed equivalent value;
It is good also as comprising the motor control apparatus provided with.

  As described above, the load torque estimated using the driving wheel acceleration and the torque command value is not always accurate. However, the error included in the inaccurate load torque can be corrected by using the idling sliding speed of the driving wheel or its equivalent value, as in an embodiment described later. Therefore, according to the first embodiment or the like, the estimated load torque at the time of slipping detection can be corrected based on the slipping speed.

  As will be described later, as the idling speed increases, the error in the load torque increases. Therefore, as in the second embodiment, the temporary return value determination step increases the correction amount as the idling speed increases. It is good also as comprising the re-adhesion control method containing these.

  Further, as in the third embodiment, the temporary return value determination step is a step of changing the correction amount according to the number of slipping detections during the re-adhesion control (for example, the continuous slipping coefficient calculation unit 85 and FIG. 6). A re-adhesion control method including processing by the multiplier 86 may be configured.

  According to the third embodiment, when idling is detected again during the temporary return, the correction amount of the load torque is changed according to the number of times, so that the temporary return value is set so as not to further idle. Can be controlled to an appropriate value.

The figure for demonstrating re-adhesion control. The figure which shows the relationship between idle running speed and a tangential force coefficient. The figure which shows the relationship between idle running speed and a tangential force coefficient correction value. The figure which shows the circuit block of the main circuit of an electric vehicle. The block diagram of a temporary return value determination part. The figure which shows the modification of a temporary return value determination part.

Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. In the following, the case where the present invention is applied to an electric vehicle will be described. However, the present invention can also be applied to an electric vehicle as long as the vehicle travels by driving a driving wheel with an electric motor (electric vehicle). Although the case where the present invention is applied to idling will be described, the present invention can be similarly applied to sliding. Needless to say, the brake control at the time of sliding can be applied not only to the regenerative brake but also to an electric vehicle that performs a brake control by a mechanical brake.
In addition, after first explaining the principle underlying the idea of the present invention, specific examples will be described in detail.

[principle]
One of the results of investigating the tangential force coefficient μ during this period from each detected value and each control value during a certain period before and after detecting idling (about 100 ms to 200 ms) (around time t1 in FIG. 1) is shown in FIG. It is. Each plot in FIG. 2 shows the tangential force coefficient μ obtained using the axial velocity V and acceleration A detected at that time at a plurality of idling sliding speeds ΔV. There is a correlation (slope) that is simplified by linear approximation in the figure that the tangential force coefficient μ decreases as the idle running speed ΔV increases.

In addition, if the tangential force coefficient in a state where no idling is generated is illustrated, μ _n in FIG. 2 is obtained. It is a tangential force coefficient corresponding to the torque pattern value τ _n ^* . Slipping when sliding occurs in this state, it drops to mu _0. That is, the tangential force coefficient μ ₀ can be said to be a tangential force coefficient when idling occurs.
In FIG. 2, the vertical axis represents the tangential force coefficient. From the relationship of load torque τ ₁ = μ × Mg × r (Mg is the axial load, r is the wheel radius), the tangential force coefficient μ and the load torque τ It said that the _l is proportional correlation tangential force coefficient μ and idling sliding velocity [Delta] V can be replaced with correlation load torque tau _l an idling sliding velocity [Delta] V.

If this correlation is used, the load torque τ _l estimated using the equation (1) is corrected to the load torque when the idle run occurs if the idle run speed ΔV before and after detecting the idle run is known. be able to.

For example, in FIG. 2, it is assumed that the idling speed when detecting idling is ΔV _S. Tangential force coefficient at this time is mu _S, also load torque tau _l estimated using equation (1), which is a value corresponding to the mu _s. However, the load torque that is originally intended to be used as a reference for the temporary torque return value τ _l ^* is the load torque when the idle running occurs, and the load torque when the idle running speed ΔV is zero. The difference between the originally desired load torque and the estimated load torque τ ₁ is an error. This error can be corrected as follows to obtain the original load torque. That is, (μ ₀ −μ _s ) is obtained from the correlation (straight line) in FIG. 2 and the idling sliding speed ΔV _s . When the torque corresponding to (μ ₀ −μ _s ) is added to the load torque τ ₁ estimated using the equation (1), the idle sliding speed ΔV when the idle running occurs is zero. The load torque can be obtained.

As described above, the estimated load torque τ _l can be corrected using the correlation between the idling sliding speed ΔV and the tangential force coefficient μ. As a specific method, as the idling sliding speed ΔV increases, the tangential force coefficient μ decreases and the correlation is constant. Therefore, the tangential force coefficient correction value Δμ corresponding to the idling sliding speed ΔV is set to, for example, FIG. Determine as follows. In FIG. 3, the correlation is set such that the tangential force coefficient correction value Δμ increases as the idling sliding speed ΔV increases.
The tangential force coefficient correction value Δμ is obtained from the idling sliding speed ΔV at the time of detecting the idling, and the obtained tangential force coefficient correction value Δμ is converted into torque and added to the load torque τ ₁ estimated by the equation (1). Thus, it is possible to obtain the load torque when the idling that is originally desired, and to set the temporary torque return value τ _l ^* to a value close to an ideal value.

[Example]
Next, an embodiment to which the above principle is applied will be described.
FIG. 4 is a diagram schematically showing a configuration related to the present embodiment in the circuit block of the main circuit of the electric vehicle, and shows one drive shaft. Although the control of the electric motor will be described below as individual control (so-called 1C1M), the applicable form of the present invention is not limited to this. For example, the present invention can be applied to 1C2M that collectively controls the two driving wheels.

  In FIG. 4, the main circuit of the electric vehicle according to the present embodiment includes an electric motor 10, a pulse generator 20, an inverter 30, a current sensor 40, and an electric motor control device 50.

The electric motor 10 is a main electric motor (main motor) that rotates the axle by being supplied with electric power from the inverter 30, and is realized by, for example, a three-phase induction motor. The pulse generator 20 is a rotation detector that detects the rotation of the drive shaft, and outputs a PG signal that is a detection signal to the vector arithmetic control device 59. Other rotation detectors such as a speed generator may be used instead of the pulse generator. The current sensor 40 is provided at the input end of the electric motor 10 and detects U-phase and V-phase currents Iu and Iv flowing into the electric motor 10. The inverter 30 is supplied with overhead power via a pantograph and a converter. Then, the output voltage is adjusted based on the voltage command values Vu ^* , Vv ^* , Vw ^* of the U-phase, V-phase, and W-phase input from the vector arithmetic control device 59, and the electric motor 10 is supplied with power.

  The electric motor control device 50 performs vector control of the electric motor 10. The electric motor control device 50 is realized by a computer or the like including a CPU, a ROM, a RAM, and the like. Configured as Further, the motor control device 50 includes a speed detection unit 51, an acceleration detection unit 53, a torque pattern value calculation device 55, a re-adhesion control device 60, a torque command determination unit 57, and a vector calculation control device 59. Yes.

  The speed detector 51 performs a filter process for smoothing the PG signal in the time axis direction to detect the axis speed V. The acceleration detection unit 53 detects the acceleration A based on the axial velocity V detected by the speed detection unit 51 that moves back and forth in time. In addition, when driving the electric motor 10 by so-called speed sensorless vector control, the speed is calculated using the driving current to the electric motor 10 detected by the current sensor 40 without using the pulse generator 20 and the PG signal. You may decide to detect an acceleration by differentiating this.

The torque pattern value calculation device 55 calculates and outputs a torque pattern value τ _n ^* of the torque command value based on the reference speed Vm and the notch signal. The reference speed Vm is a traveling speed of a train, and may be a speed obtained from a driver's cab, for example, or may be an axial speed of a follower wheel of a T car. Further, among the shaft speeds of the respective axes in the vehicle, the minimum value may be determined during power running, and the maximum value or the like may be determined during braking. Since the calculation process of the torque pattern value τ _n ^* is known, the description thereof is omitted.

The torque command determination unit 57 determines the torque pattern value τ _n ^* output from the torque pattern value calculation device 55 as the torque command value τ _e ^{* in the} normal time when idling is not detected, and during re-adhesion control is being executed. The re-adhesion command value τ _r ^* output from the re-adhesion control device 60 is determined as the torque command value τ _e ^* . Whether or not the re-adhesion control is being executed is determined by a control signal (not shown) from the re-adhesion control device 60.

The vector arithmetic control unit 59 is a d-axis component excitation current component Id and a q-axis component torque current component (motor torque) obtained by converting the Iv and Iu detected by the current sensor 40 by dq axis coordinates. The voltage command values Vu ^* , Vv ^* , and Vw ^* for the inverter 30 are generated based on the component current (Iq), the torque command value τ _e ^{*, and the} like. Note that the arithmetic processing for calculating the voltage command values Vu ^* , Vv ^* , Vw ^* is a known arithmetic processing, and a description thereof will be omitted.

The re-adhesion control device 60 includes an idling / sliding detection unit 61, a recovery detection unit 63, a re-adhesion detection unit 65, and a re-adhesion control unit 67. The general operation of the re-adhesion control device 60 is the same as the re-adhesion control described with reference to FIG. That is, when the idle running detection unit 61 detects the occurrence of idle running using the reference speed Vm, the shaft speed V, and the acceleration A, the re-adhesion control unit 67 starts the re-adhesion control and sets the torque pattern value τ _n ^* . The command value gradually reduced is generated and output as the re-adhesion command value τ _r ^* . In addition, a re-adhesion control start signal is output to the torque command determination unit 57.

Then, when the torque is reduced and the recovery detection unit 63 detects the start of recovery from the idle running to the original adhesion running, the re-adhesion control unit 67 stops the reduction of the re-adhesion command value τ _r ^*. Hold.
Thereafter, when the re-adhesion detection unit 65 detects re-adhesion, the re-adhesion control unit 67 performs temporary return control that gradually increases the re-adhesion command value τ _r ^* so as to approach the temporary torque return value τ _l ^*. . Then, if the idling / sliding detector 61 does not detect for a predetermined time, it is determined that the temporary return is completed, and the re-adhesion control unit 67 changes the re-adhesion command value τ _r ^* to the torque pattern value τ _n ^* . When the values gradually approach and coincide with each other, the re-adhesion control is completed, and a completion signal is output to the torque command determination unit 57.

FIG. 5 is a diagram illustrating a functional block of the temporary return value determination unit 70 that is a processing unit that determines the temporary torque return value τ _l ^* .
The temporary return value determination unit 70 includes a load torque estimation unit 71, a load torque holding unit 72, and a correction unit 80. The load torque estimating unit 71 estimates the load torque according to the equation (1) using the acceleration A detected by the acceleration detecting unit 53 and the torque pattern value τ _n ^* calculated by the torque pattern value calculating device 55, The torque is output to the torque holding unit 72.
The load torque holding unit 72 updates and holds the held load torque to the load torque input from the load torque estimating unit 71 at the timing when the idling / sliding detection signal is input from the idling / sliding detecting unit 61. Therefore, the load torque when the occurrence of idling is detected by the load torque estimating unit 71 and the load torque holding unit 72 is estimated.

  The correction unit 80 is a functional unit that corrects the load torque when the occurrence of idling is detected, which is held by the load torque holding unit 72, and includes a subtractor 81, an idling running speed holding unit 82, and a tangential force. A coefficient correction value calculation unit 83, a conversion unit 87, and an adder 89 are included.

  The subtractor 81 calculates an idle running speed (speed difference) ΔV by subtracting the reference speed Vm from the shaft speed V detected by the speed detecting unit 51, and outputs it to the idle running speed holding unit 82. The idle running speed holding unit 82 updates the held idle running speed to the idle running speed input from the subtracter 81 at the timing when the idle running detection signal is input from the idle running detection unit 61. . Therefore, the subtractor 81 and the idle running speed holding unit 82 calculate and hold the idle running speed ΔV when the occurrence of the idle running is detected.

  The tangential force coefficient correction value calculation unit 83 stores data for obtaining the tangential force coefficient correction value Δμ from the idling sliding speed ΔV shown in FIG. 3, and sets the idling sliding speed held in the idling sliding speed holding unit 82. The corresponding tangential force coefficient correction value is calculated and output to the conversion unit 87. The data for obtaining the tangential force coefficient correction value Δμ from the idling sliding speed ΔV may be function formula data or table format data.

  The conversion unit 87 is a functional unit that converts tangential force into torque, and a conversion formula is set in advance based on known values such as shaft weight, wheel diameter, gear ratio, and the like. The conversion unit 87 converts the tangential force coefficient correction value Δμ calculated by the tangential force coefficient correction value calculation unit 83 into torque, and outputs the torque to the adder 89 as a load torque correction value for correcting the load torque.

The adder 89 performs correction by adding the load torque correction value output from the conversion unit 87 to the load torque held by the load torque holding unit 72 when the occurrence of idling is detected. The subsequent load torque is generated and output as a temporary torque return value τ _l ^* .

According to the above embodiment, the error included in the estimated load torque is reduced by correcting the load torque estimated when the occurrence of idling is detected based on the idling speed when the idling is detected. Thus, a value close to the load torque when the original idling occurs can be obtained. When the torque temporary return value τ _l ^* is determined without correcting the load torque estimated when the occurrence of idling is detected, the value is lower than the original value. According to the present embodiment, the temporary torque return value τ _l ^* can be brought close to the original value. As a result, it is possible to increase the traction force of the electric vehicle during the stage return control of the idling / sliding control as compared with the conventional case.

[Modification]
The embodiments to which the present invention can be applied are not limited to the above-described embodiments. For example, re-adhesion control unit 67 generates a re-adhesive for instruction value tau _r ^*, the torque command value and the torque command determiner 57 switches the Torukupatan value tau _n ^* re tackifying command value τ _r ^* τ _e ^* Explained as determining. The re-adhesion control unit 67 does not generate the re-adhesion command value τ _r ^* , but calculates a reduction amount (or reduction ratio) with respect to the torque pattern value τ _n ^* , and the torque command determination unit 57 calculates the torque pattern. The torque command value τ _e ^* may be determined by subtracting the torque corresponding to the reduction amount from the value τ _n ^* .

It is also possible to employ a method of Equation (2) to determine the temporary return value tau _l ^* torque by multiplying a constant factor k _D to the load torque corrected by the correction unit 80 of FIG.

In addition, idling may be detected again during the re-adhesion control. In that case, the temporary torque return value τ _l ^* may be reduced in accordance with the number of times of idling (the number of re-detections). FIG. 6 is a diagram illustrating a modification of the temporary return value determination unit in this case. 6 is different from the temporary return determination unit 70 in FIG. 5 in the correction unit 80A. In addition, a multiplier 78 is further added in order to employ the method of Expression (2).

The correcting unit 80A has a configuration in which a continuous idling sliding coefficient calculating unit 85 and a multiplier 86 are further added to the correcting unit 80 of FIG. The continuous idle running coefficient calculation unit 85 calculates the continuous idle running coefficient k _C according to the number of occurrences of the idle running in one re-adhesion control using the following equation (3).
k _C = 1− (i−1) × C _C (3)
Here, i is the number of continuous idling and _CC is a reduction factor. The continuous idling / sliding coefficient calculating unit 85 has a counter circuit, counts up every time a detection signal is input from the idling / sliding detection unit 61, and is reset by a detection signal from the re-adhesion detection unit 65. The number of continuous idling times i is counted. Reduction factor C _C is defined a value of less than 1 greater than zero, preferably 0.1 or less value greater than zero is defined to.
Continuous idle coefficient calculation unit 85, in the case of a continuous idle sliding number i is 1, and 1 continuous idle sliding coefficient k _C, since, according to the continuous slipping sliding skid number i increases, the continuous slipping sliding coefficient k _C Decrease gradually.

The multiplier 86 multiplies the tangential force coefficient correction value Δμ calculated by the tangential force coefficient correction value calculation unit 83 by the continuous idling sliding coefficient k _C calculated by the continuous idling sliding coefficient calculation unit 85 to obtain a conversion unit. Output to 87. As a result, every time the occurrence of idling is detected during the re-adhesion control, the amount of correction of the load torque becomes small, and the recurrence of idling can be suppressed.

The temporary return value determination unit 70A includes a multiplier 78 as a configuration for realizing the expression (2). The multiplier 78, the load torque after being corrected by the correction unit 80A, multiplies the coefficient k _D. It coefficient _{k D} can also be defined as a value of, for example, about 0.8 to 0.9, it is also possible to 1.0. In the case of 1.0, the configuration is the same as that of FIG. The value after multiplication by the multiplier 78 is the torque temporary return value τ _l ^* .

  In the above-described embodiment, the load torque estimated when the occurrence of idling is detected is corrected using the idling speed when the idling is detected. Specifically, it has been described that the tangential force coefficient correction value Δμ is calculated based on the idle running speed ΔV when the idle running is detected, and the estimated load torque is corrected. However, the estimated load torque may be corrected using an equivalent value that can be handled equivalently to the idling sliding speed ΔV. For example, even if the horizontal axis in FIGS. 2 and 3 is the “slip rate”, the relationship between the tangential force coefficient μ and the tangential force coefficient correction value Δμ can be said to be the same, so the slip rate is used instead of the idling sliding speed ΔV. The estimated load torque may be corrected. In this case, the idling sliding speed holding unit 82 holds the slip ratio instead of the idling sliding speed ΔV, and the tangential force coefficient correction value calculating unit 83 calculates the tangential force coefficient correction value Δμ from the slip ratio.

  Further, in the above-described embodiment, the correlation between the idle running speed ΔV and the tangential force coefficient μ before and after detecting the idle running is regarded as constant (see FIG. 2), and the idle running speed ΔV is based on this fixed relation. The tangential force coefficient correction value [Delta] [mu] corresponding to is determined (see FIG. 3). However, this correlation may be updated as needed. Specifically, after detecting slipping, the slipping speed ΔV and the tangential force coefficient μ (more specifically, various quantities for obtaining the tangential force coefficient) are sampled a plurality of times to obtain a correlation. For example, the number of samplings may be set to 2, and the linear approximation correlation (gradient) as shown in FIG. 2 may be obtained. Then, based on this correlation, for example, the sign of the slope is reversed, and the direct proportional relationship between the idling sliding speed ΔV and the tangential force coefficient correction value Δμ as shown in FIG. 3 may be updated and defined.

DESCRIPTION OF SYMBOLS 50 Electric motor control apparatus 55 Torque pattern value calculating apparatus 57 Torque command determination part 60 Re-adhesion control apparatus 61 Idling sliding detection part 67 Re-adhesion control part
70 Temporary return value determination unit
71 Load torque estimation unit
72 Load torque holding part
80 Correction section
81 Subtractor
82 idle running speed holding part
83 Tangential force coefficient correction value calculation unit
87 Conversion part
89 Adder

Claims (4)

When the occurrence of idling or sliding of a driving wheel (hereinafter referred to as “idling / sliding”) is detected, the torque command value is lowered, temporarily returned to the original return value, and then returned to the original torque command value. A re-adhesion control method for re-adhering the driving wheel by controlling a driving motor,
A load torque estimation step of estimating a load torque at the time of idling detection using the acceleration relating to the rotation of the driving wheel and the torque command value;
A temporary return value determining step of determining the temporary return value by correcting the estimated load torque using an idle running speed of the driving wheel or a value corresponding to the idle running speed;
Re-adhesion control method including. The temporary return value determining step includes a step of increasing a correction amount as the idling sliding speed is increased.
The re-adhesion control method according to claim 1. The temporary return value determination step includes a step of changing the correction amount according to the number of idling sliding detections during the re-adhesion control.
The re-adhesion control method according to claim 1 or 2. When the occurrence of idling of the driving wheel is detected, the torque command value is lowered, and after returning to the temporary return value, the motor is driven by controlling the motor that drives the driving wheel by the step return control to return to the original torque command value. A motor control device for re-adhesion,
Load torque estimating means for estimating a load torque at the time of idling detection using the acceleration relating to the rotation of the driving wheel and the torque command value;
A temporary return value determination means for determining the temporary return value by correcting the estimated load torque using an idle running speed of the driving wheel or a value corresponding to the idle running speed;
An electric motor control device.

Images