JP7136459B2 - meshing clutch - Google Patents

Description

The present invention relates to dog clutches.

Conventionally, there has been known a technique for detecting the rotational phases of two clutch members and meshing engagement teeth without colliding with each other. For example, in Patent Document 1, Hall sensors are used as means for detecting the rotational phase, and separate Hall sensors are provided for two engaging teeth, or one Hall sensor is provided so as to straddle two engaging teeth. For example, a technique has been proposed in which the rotation phase difference and the rotation speed difference between both clutch members are detected from the AC component of the sensor output, and the engagement timing of the clutch members is controlled based on these detected values.

Japanese Patent Publication No. 2013-513766

By the way, since a dog clutch has a mechanical or electrical operation delay time, it is necessary to engage the clutch members at a timing that takes into consideration the operation delay time of the clutch in order to engage the engagement teeth without colliding with each other. There is In addition, the operation delay time of the clutch is not always constant, and changes according to the configuration and operating conditions of the power transmission system. Therefore, it is necessary to control the clutch in consideration of such circumstances.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a meshing clutch that allows two clutch members to be meshed at an appropriate timing in consideration of clutch operation delay time.

In order to solve the above problems, the dog clutch of the present invention includes a first clutch member (11) having a plurality of first engagement teeth (13) arranged in the circumferential direction, and a releasable clutch member (11) on the first engagement teeth. A second clutch member (12) having a plurality of meshing second engagement teeth (14) arranged in the circumferential direction, the first clutch member and the second clutch member are moved relative to each other in the clutch axial direction to achieve the first engagement. An actuator (5) for meshing the mating tooth and the second engagement tooth, a detection device (6) for detecting a rotational phase difference between the first clutch member and the second clutch member, and an actuator based on the detected rotational phase difference. a control device (7) for outputting an operation command to , the control device obtains the engagement timing of the first clutch member and the second clutch member from the detected rotational phase difference, and calculates at least based on the past engagement timing One future engagement timing is predicted, and an operation command is output to the actuator when the difference between the future engagement timing and the current time reaches the clutch actuation delay time.

According to the above configuration, by starting the meshing operation at a timing that is earlier than the predicted value of the future meshing timing by the clutch operation delay time, the two clutch members are properly meshed without causing the engagement teeth to collide with each other. be able to. Further, by correcting the predicted value of the future engagement timing with the correction value using the rotational speed difference at the time of execution of the prediction, it is possible to reduce the prediction error with relatively simple processing.

1 is a schematic diagram of a dog clutch showing a first embodiment of the present invention; FIG. In the first embodiment, (a) is a schematic diagram showing the detection range of the detection device, and (b) is a waveform diagram of an area detection signal output by the detection device. FIG. 5 is a characteristic diagram showing a procedure for predicting future engagement timing from past engagement timing and correcting the predicted value in the first embodiment; In the first embodiment, (a) is a characteristic diagram showing changes in the number of rotations of a dog clutch for a vehicle, and (b) is a waveform diagram showing a procedure for predicting future engagement timing. FIG. 4 is a characteristic diagram showing a procedure for correcting predicted values of several future engagement timings in the first embodiment; FIG. 9 is a characteristic diagram showing a procedure for obtaining an engagement timing by binarizing an area detection signal waveform in the second embodiment of the present invention; In the second embodiment, (a) is a waveform diagram including engagement timing and anti-phase timing, and (b) is a characteristic diagram showing a procedure for predicting future engagement timing from both timings. FIG. 11 is a characteristic diagram showing a procedure for obtaining engagement timing from an integrated value of envelope curves in the third embodiment of the present invention; FIG. 10 is a waveform diagram showing a procedure for determining engagement timing from an amplitude decrease speed or an amplitude increase speed in the fourth embodiment of the present invention; FIG. 11 is a waveform diagram showing a procedure for determining engagement timing by Hilbert transform processing in the fifth embodiment of the present invention; In the fifth embodiment, (a) is a characteristic diagram showing re-Hilbert transformation processing, and (b) is a characteristic diagram showing a procedure for predicting future engagement timing from the phase difference after re-Hilbert transformation. FIG. 11 is a schematic diagram showing a configuration for predicting future engagement timing by FFT processing in a sixth embodiment of the present invention; In the seventh embodiment of the present invention, (a) is a schematic diagram showing a rotation speed change due to disturbance, and (b) is a schematic diagram showing a meshing error due to disturbance.

Hereinafter, dog clutches according to a plurality of embodiments of the present invention will be described based on the drawings. In addition, in each embodiment, the same code|symbol is attached|subjected to the substantially same component, and description is abbreviate|omitted.

<First Embodiment>
As shown in FIG. 1, the dog clutch 1 of the first embodiment includes a first clutch member 11 on the drive side and a second clutch member 12 on the driven side in the direction in which the clutch axis A extends (hereinafter abbreviated as the axial direction). It has The first clutch member 11 is rotated via a drive shaft 3 by a power device 2 such as a motor. The second clutch member 12 is engaged with the first clutch member 11 and transmits the power of the power unit 2 to a driven member (not shown) via the driven shaft 4 .

A plurality of first engagement teeth 13 and a plurality of second engagement teeth 14 are formed on the opposing end surfaces of the first clutch member 11 and the second clutch member 12 over the entire circumference of the clutch members. The engaging teeth 13 and 14 are formed in an uneven shape that releasably mesh with each other. The first clutch member 11 and the second clutch member 12 are moved relative to each other in the axial direction by the actuator 5, and the engagement position where the first engagement tooth 13 and the second engagement tooth 14 are engaged and the release position where they are separated. and

A detection device 6 for detecting a rotational phase difference between the first clutch member 11 and the second clutch member 12 is installed perpendicularly to the clutch axis A in the vicinity of the outer periphery of the first clutch member 11 and the second clutch member 12 . As shown in FIG. 2(a), the detection device 6 has a detection range 16 that partially includes one first engaging tooth 13 and one second engaging tooth 14. there is Then, the detection device 6 outputs a signal that changes according to the areas of the engagement teeth 13 and 14 included in the detection range 16 to the control device 7 as a signal indicating the rotational phase difference between the clutch members 11 and 12 . As the detection device 6, for example, a distance sensor that detects an area from distance measurement data, a camera that can detect an area by image processing of imaging data, or the like can be used.

The signal output by the detection device 6 is a composite wave of two waveforms corresponding to the rotation phase and rotation speed of each of the first clutch member 11 and the second clutch member 12 . When there is a rotational speed difference between the clutch members 11 and 12, this composite wave becomes a waveform accompanied by a beat as shown in FIG. . For this reason, the control device 7 determines when the amplitude of the signal output by the detection device 6 becomes below a certain value or becomes 0, and outputs an operation command, that is, a meshing command to the actuator 5 at a timing earlier by the clutch operation delay time. If so, it should be possible to avoid collisions between the engaging teeth.

However, the operation delay time of the dog clutch 1 is caused by delay in energization of the actuator 5, inertia of the clutch members 11 and 12, mechanical friction, axial movement speed, and the like. to the actual meshing of the engaging teeth 13 and 14 is included. Therefore, when there is a rotational speed difference, if an operation command is issued at or after the timing at which meshing is possible, the timing at which the engaging teeth 13 and 14 actually mesh is included in the period when meshing is not possible, and as a result, Engaging teeth may collide with each other.

Therefore, the control device 7 of the first embodiment obtains and stores the engagement timing of the engagement teeth 13 and 14 from the rotational phase difference, and predicts and predicts the future engagement timing based on a plurality of past engagement timings. An operation command is output to the actuator 5 when the difference between the future engagement timing and the current time reaches the clutch operation delay time. Specifically, as shown in FIG. 3, the control device 7 stores at least two or more past engagement timings (Tp), calculates an approximate line passing through each point, and future engagement timing (Tf) is predicted, and a correction value (Cv) corresponding to the rotational speed difference is added to the predicted value.

Correction processing is performed to reduce errors in the predicted values. In general, the cycle of engagement timing changes according to the difference in rotational speed between the two clutch members 11 and 12 . For example, in a dog clutch used in a power transmission system for a vehicle, as shown in FIG. In this case, as the rotation speed of the motor approaches the rotation speed of the axle, the difference in rotation speed between the first clutch member 11 on the motor side and the second clutch member 12 on the axle side gradually decreases, and the engagement timing cycle. gradually lengthens.

A periodic change in meshing timing due to a rotational speed difference can be predicted by an approximation formula. Specifically, as shown in FIG. 4(b), the control device 7 stores a plurality of past engagement timings Tp based on the area detection signal from the detection device 6 in a storage area, and an approximation formula is calculated, and an approximation formula can be used to estimate the future engagement timing (Tf) after the current time. At this time, if a higher-order formula or exponential-logarithmic formula is used, an approximation formula with higher prediction accuracy can be obtained. have an adverse effect on Therefore, it is conceivable to reduce the influence of the prediction error by adopting a low-order approximation formula such as a quadratic formula, for example.

In order to reduce the prediction error, the control device 7 of this embodiment makes a prediction using a low-order approximation formula and performs correction using the rotation speed difference when the prediction is executed. Referring to FIG. 4(a) again, when the control is performed so that the rotation speed of the motor is brought closer to the axle rotating at a certain speed, the change in the rotation speed of the motor is determined by the control characteristics of the motor. The time change of the difference is determined for each product. Therefore, by grasping the control characteristics of the motor to be used, the rotational speed difference at the time of prediction execution is used to estimate the rotational speed difference at future meshing timing, and the meshing timing is predicted based on the estimated rotational speed difference. values can be corrected.

In this embodiment, the error between the predicted future engagement timing and the actual engagement timing is totaled according to the rotational speed difference at the time of execution of the prediction, and a correction formula is created in advance. Here, as shown in FIG. 5, by predicting the future engagement timing (Tf) not only one time ahead, but also two, three or more times ahead, is shorter than the clutch actuation delay time, it is possible to perform correction processing such as using the predicted value two times later. In this case, it is necessary to prepare the above-mentioned correction formulas for a required number of times, such as one time ahead and two times ahead.

Therefore, according to the meshing clutch 1 of the first embodiment, the future meshing timing (Tf) is predicted from the past meshing timing (Tp), and the meshing operation is started at a timing earlier than the predicted value by the clutch operation delay time. By doing so, the two clutch members 11 and 12 can be meshed smoothly without causing the engaging teeth 13 and 14 to collide with each other. In addition, since the predicted value of the future engagement timing (Tf) is corrected using the difference in the number of rotations of the two clutch members 11 and 12 at the time when the predicted value is obtained, the prediction error can be corrected by relatively simple processing. can be reduced.

By the way, in the first embodiment, the control device 7 obtains the phase at which the waveform of the area detection signal from the detection device 6 (see FIG. 2) becomes a node (the phase at which the amplitude is equal to or less than a predetermined value or 0) as the engagement timing. However, the invention is not limited to this. In some embodiments exemplified below, the control device 7 obtains a timing that can be meshed by a procedure different from that of the first embodiment.

<Second embodiment>
In the second embodiment shown in FIGS. 6 and 7, the control device 7 performs processing for binarizing the waveform of the area detection signal. That is, a certain threshold is set for the output of the detection device 6, and the output waveform is differentiated into periods above the threshold and periods below the threshold. Since the amplitude of the output waveform becomes small near the timing at which meshing is possible, a period below the threshold occurs. On the other hand, even at times other than the timing at which meshing is possible, depending on the amplitude of the output, a period below the threshold occurs. become Then, a period (P1, P2, P3) of a predetermined length or longer with a discrimination value of 0 is determined to be the timing at which meshing is possible, and the midpoint from the start to the end of that period is determined as the meshing timing (T1, T2, T3). and stored in the storage area.

Further, as shown in FIG. 7(a), the halfway point from the start to the end of the period (Px) longer than a predetermined time period out of the periods exceeding the threshold value is set to the opposite phase timing with respect to the preceding and succeeding engagement timings (T2, T3). It can also be obtained as (Tx). In this way, as shown in FIG. 7B, the past anti-phase timings (Tx1, Tx2) are added to the past engagement timings (Tp1 to Tp3) used for prediction, and based on a greater number of past timing data, Therefore, the future engagement timing (Tf1) can be predicted with high accuracy.

<Third Embodiment>
In the third embodiment shown in FIG. 8, the control device 7 obtains the past engagement timing according to the integration result of the output peak value of the detection device 6. FIG. That is, the absolute value of the peak of the area detection signal output by the detection device is integrated, or the absolute value of the output envelope is integrated, and the integrated value is plotted along the time axis to form a curve as shown in the figure. get The phase at which the gradient of this curve becomes 0, that is, the time point at which the first derivative of the time change of the integrated value becomes 0, is determined as the engagement timing (T1, T2, T3). The point in time when the second-order differential value of the time change of the value becomes 0 is obtained as the opposite phase timing (Tx1, Tx2, Tx3) described above.

<Fourth Embodiment>
In the fourth embodiment shown in FIG. 9, the controller 7 predicts the next phase at which the amplitude of the area detection signal decreases from the rate at which the amplitude of the area detection signal decreases as the future engagement timing. Here, the rate of decrease in amplitude is the same as the slope of the envelope. However, since this method can only predict the future one time ahead, it cannot be applied when the clutch activation delay time is longer than the time from the current time to the prediction timing. Therefore, under the condition that the rotational speed difference does not change, the amplitude increase speed is equal to the amplitude decrease speed, so it is possible to predict the next engagement timing from the speed of the amplitude increase interval before the amplitude decrease interval.

<Fifth Embodiment>
In the fifth embodiment shown in FIG. 10, the control device 7 performs Hilbert transform on the waveform of the area detection signal to numerically obtain the engagement timing. In the Hilbert transform, if the output of the detector 6 is g(t) and the post-Hilbert transform is h(t), the output envelope f(t) is
f(t)=±√(h(t) ² +g(t) ² )
is represented by Therefore, the time when the envelope curve falls below the threshold value, in the example shown in FIG. 10, the time when the calculated value of the envelope curve becomes the output median value can be obtained as the engagement timing (T).

Also, as shown in FIG. 11(a), if the envelope f(t) is Hilbert-transformed again to obtain the value k(t), the envelope at the current time is obtained from arctan(k(t)/f(t)) The phase of the line f(t), that is, the rotational phase difference between the clutch members 11 and 12 can be obtained.Therefore, as shown in FIG. can improve the prediction accuracy of

<Sixth Embodiment>
In the sixth embodiment shown in FIG. 12, the control device 7 has an FFT circuit 41 and fast Fourier transforms the area detection signal to numerically obtain the engagement timing. According to the fast Fourier transform, since the rotational phases of the two clutch members 11 and 12 can be obtained separately, the control device 7 can easily obtain the engagement timing, the opposite phase timing, and other arbitrary timings. The actuator 5 can be variously controlled according to predicted values based on past data.

For example, by performing a fast Fourier transform on the area detection signal, the rotational phase θ and the angular velocity ω of the clutch members 11 and 12 can be obtained at the same time, and the output can be mathematically expressed as follows to predict the future engagement timing. It is possible. In the following equations, P is the amplitude of the area detection signal.

Signal from first clutch member 11: X1=P×cos(ω1×t+θ1)
Signal from second clutch member 12: X2=P×cos(ω2×t+θ2)
Output of detector 6:
Y=X1+X2
=P×cos(ω1×t+θ1)+P×cos(ω2×t+θ2)
=2×P×cos(((ω1+ω2)t+(θ1+θ2))/2)×
cos(((ω2−ω1)t+(θ2−θ1))/2)

In the above formula, the cos part using the sum of ω and θ is the high frequency component of the output, and the cos part using the difference between ω and θ is the low frequency part represented by the envelope. Since the timing at which this low-frequency portion becomes 0 is the engagement timing, if ω3=ω1-ω2 and θ3=θ1-θ2, the engagement timing after N times is as follows.
(ω3t+θ3)/2=(N+1/2)π
By solving this for t, the engagement timing t N times ahead is given by
t=((π−θ3)+2Nπ)/ω3
more predictable.

<Seventh Embodiment>
In each of the above-described embodiments, the procedure for predicting the future meshing timing has been described. However, when the rotation speed of the clutch members 11 and 12 is disturbed, a time lag occurs between the predicted meshing timing and the actual meshing timing. . For example, in a dog clutch for a vehicle, when the first clutch member 11 is connected to the motor and the second clutch member 12 is connected to the axle, as shown in FIG. As shown in FIG. 13(b), when the vehicle runs over a step S on the road surface, the rotational speed of the axle may momentarily lag behind the rotational speed of the motor. If this delay occurs between obtaining the predicted value and operating the actuator, a meshing error occurs.

Here, when the number of revolutions on the axle side is high, the actual engagement timing is slightly delayed from the predicted timing due to the above disturbance. Therefore, even if the engagement teeth collide with each other, it is possible to engage immediately afterward due to slippage. be. However, when the rotational speed of the axle is small, the actual engagement timing becomes earlier than the predicted timing due to disturbance, so the engaged teeth collide with each other, and the engagement is delayed until the next engagement timing is shifted by one tooth. Because there is not, the delay becomes large.

Therefore, in the seventh embodiment, when the axle-side rotation speed is small, the control device 7 outputs an operation command to the actuator from the beginning by the delay time due to the disturbance prior to the predicted timing, and the engagement timing is delayed due to the disturbance. To control the engaging teeth 13 and 14 so that they mesh with each other without colliding even when an error occurs. If there is no disturbance, the engaging teeth will collide with each other, but the delay until the actual meshing is only the amount of advance, so the maximum delay is made smaller than when waiting for the engaging tooth to shift by one tooth. be able to.

In each of the above-described embodiments, the control device 7 executes one algorithm in order to predict future engagement timing, but the control device 7 may switch and execute a plurality of algorithms depending on the situation. It is possible. Specifically, a plurality of prediction units may be provided in the control device 7, and each prediction unit may be configured to switch between different algorithms, such as Hilbert transform and fast Fourier transform. In addition, the present invention is not limited to the above embodiments, and it is possible to arbitrarily change the shape and configuration of each part without departing from the gist of the invention.

DESCRIPTION OF SYMBOLS 1... Mesh clutch, 5... Actuator, 6... Detection device,
7...control device, 11...first clutch member, 12...second clutch member,
13... First engagement tooth, 14... Second engagement tooth,
Cv: correction value, T: engagement timing,
Tp: past engagement timing, Tf: future engagement timing.

Claims (13)

a first clutch member (11) having a plurality of first engagement teeth (13) arranged in a circumferential direction;
a second clutch member (12) in which a plurality of second engagement teeth (14) releasably meshing with the first engagement teeth are arranged in a circumferential direction;
an actuator (5) that relatively moves the first clutch member and the second clutch member in the clutch axial direction to mesh the first engagement tooth and the second engagement tooth;
a detection device (6) for detecting a rotational phase difference between the first clutch member and the second clutch member;
a control device (7) that outputs an operation command to the actuator based on the detected rotational phase difference,
The control device obtains engagement timings of the first clutch member and the second clutch member from the detected rotational phase difference, predicts at least one future engagement timing based on past engagement timings, and predicts future engagement timings. A dog clutch configured to output an operation command to the actuator when the difference between the timing and the current time reaches a clutch operation delay time. 2. A meshing clutch according to claim 1, wherein said control device adds a correction value according to a difference in rotational speed between said first clutch member and said second clutch member to a predicted value of future meshing timing. The detection device has a detection range (16) of a predetermined area, and detects the first clutch member and the first clutch member by a signal corresponding to the areas of the first engagement tooth and the second engagement tooth included in the detection range. 3. A dog clutch according to claim 1, wherein a rotational phase difference of said second clutch member is detected. 4. A dog clutch according to claim 3, wherein said control device stores a midpoint of a period in which the output of said detection device falls below a threshold value as the engagement timing of said first clutch member and said second clutch member. The control device stores a midpoint of at least one period in which the output of the detection device exceeds a threshold value as opposite phase timing at which the first engagement tooth and the second engagement tooth face each other. 5. A dog clutch as set forth in claim 4, wherein future engagement timing is predicted based on the timing and at least one past anti-phase timing. The control device integrates the output of the detection device, and stores the point in time when the first differential value of the time change of the integrated value becomes 0 as the engagement timing of the first engagement tooth and the second engagement tooth. , the point in time when the second derivative of the time change of the integrated value becomes 0 is stored as opposite phase timing at which the first engagement tooth and the second engagement tooth face each other, and a plurality of past engagement timings and at least one 3. A meshing clutch according to claim 1, wherein future meshing timing is predicted based on past anti-phase timing. 4. A dog clutch according to claim 3, wherein said control device predicts a time point at which the speed of decrease in amplitude of the signal output from said detection device becomes 0 as a future engagement timing. 4. The meshing according to claim 3, wherein the control device estimates a subsequent amplitude decreasing speed from the amplitude increasing speed of the signal output from the detecting device, and predicts a time when the amplitude decreasing speed becomes 0 as a future meshing timing. clutch. The control device obtains an envelope of the output from a signal obtained by Hilbert transforming the output of the detection device, and stores the point in time when the envelope falls below a threshold value as the engagement timing of the first clutch member and the second clutch member. Item 4. The dog clutch according to Item 3. The control device obtains an envelope of the output from a signal obtained by Hilbert-transforming the output of the detection device, obtains a phase by performing Hilbert-transformation on the envelope again, sets the phase at which the engagement timing is 0, and calculates other phase points. 4. A dog clutch according to claim 3, which is used for prediction of future engagement timing. The control device performs a fast Fourier transform on the output of the detection device to obtain rotational phases of the first clutch member and the second clutch member, and based on the rotational phase difference between the first clutch member and the second clutch member. 4. The dog clutch of claim 3, which stores the engagement timing of the clutch members. When the number of revolutions of either the first clutch member or the second clutch member is lower than the number of revolutions of the other clutch member due to disturbance, the control device controls the delay time due to the disturbance from the predicted engagement timing. A dog clutch according to any one of claims 1 to 11, wherein an operation command is output to said actuator at a timing preceding by . 2. The dog clutch according to claim 1, wherein said control device has a plurality of prediction units for predicting future engagement timing, and switches between said prediction units.

Images