JP2007112350A - Transmission controller - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To highly precisely achieve the learning of engagement pressure control for the rotary member of a transmission without generating cost increase in a hybrid power unit for performing the power synthesis of partial power sources through the transmission for shifting gears by clutch-to-clutch processing. <P>SOLUTION: When it is detected that the hybrid power unit for a vehicle is put in a coast traveling status, and is down-shifted (S100: yes), a motor generator functions as a motor to control the outputs of partial power sources to a fixed torque (S112, S118). The engagement pressure at the release clutch side of the transmission is gradually reduced in this period, thereby the start timing of an inertia phase can be generated. The deviation on control is highly precisely reflected on the deviation of the start timing of the inertial phase. Therefore, the learning of the pressure control at the release clutch side is executed based on the adjustment history of the engagement pressure in the start timing of the inertial phase. Thus, it is possible to achieve highly precise learning without generating cost increase due to torque sensor installation or the like. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention combines the power of a plurality of power sources and outputs them to a rotary output shaft, and the power synthesis of some power sources is performed through a transmission that shifts by clutch-to-clutch processing. The present invention relates to a machine control device.

  As a hybrid power device, for example, a hybrid power device for a vehicle, there is known a device that combines the power of an internal combustion engine and the power of an electric motor and outputs the combined power to a rotary output shaft (see, for example, Patent Document 1). The electric motor in this case combines the power with the power of the internal combustion engine with respect to the rotation output shaft through the transmission. By shifting with this transmission, the electric motor gives an appropriate rotation and torque to the rotation output shaft.

Technology for correcting the initial hydraulic pressure of the disengagement clutch by learning the amount of change in the member rotation speed at the beginning of the inertia phase when downshifting at power-on in hydraulic control of an automatic transmission that shifts by clutch-to-clutch processing Has been proposed (see, for example, Patent Document 2).
Japanese Patent Laying-Open No. 2004-203219 (page 8-10, FIG. 1) Japanese Patent Laid-Open No. 11-108168 (pages 4-6, FIGS. 3, 4 and 9)

For the hybrid power unit as shown in Patent Document 1, it is desired to execute the learning control as in Patent Document 2 for high-precision shift control.
However, the technique of Patent Document 2 is a technique for executing learning by calculating a transmission input torque at power-on. In such a learning situation, the change of the transmission input torque accompanying the accelerator operation is large, the accuracy of the estimation calculation of the transmission input torque is low, and the improvement of the learning accuracy is limited. Further, if an accurate input torque is to be obtained, a torque sensor is used, resulting in an increase in cost.

  The present invention provides a hybrid power unit that performs power synthesis of some power sources via a transmission that shifts by clutch-to-clutch processing, and controls engagement pressure control for a rotating member of the transmission without incurring an increase in cost. The purpose is to perform learning with high accuracy.

In the following, means for achieving the above object and its effects are described.
The transmission control apparatus according to claim 1 synthesizes the powers of a plurality of power sources and outputs them to the rotation output shaft, and the power synthesis of a part of the power sources is a transmission that shifts by clutch-to-clutch processing. A transmission control device for a hybrid power unit that is configured to determine whether the transmission is shifting or not, and when the shifting determination unit determines that the shifting is in progress, A constant torque control means for forming a constant torque control period for controlling the output of a part of the power source to a constant torque, and a release clutch of the transmission within the constant torque control period formed by the constant torque control means; Release-side engagement pressure gradual reduction means for gradually reducing the engagement pressure on the side in a reference pressure pattern, and start timing of inertia phase generated during a gradual decrease period in the reference pressure pattern by the release-side engagement pressure gradual reduction means Inertia phase start timing detection means for detecting the release, and the release based on the engagement pressure adjustment history by the release side engagement pressure gradual reduction means up to the start timing detected by the inertia phase start timing detection means Disengagement-side pressure control learning means for performing clutch-side pressure control learning is provided.

  The constant torque control means forms a constant torque control period for controlling the output of some power sources to a constant torque at the time of shifting. As a result, regardless of the state of other power sources, the disengagement clutch side rotating member of the transmission receives a constant stable torque directly or indirectly from some power sources during the constant torque control period.

  Then, the release side engagement pressure gradual reduction means gradually reduces the engagement pressure on the release clutch side of the transmission with the reference pressure pattern within this constant torque control period. As a result, the release clutch side rotation member that was not initially rotated starts to rotate at the timing when the frictional force due to the engagement pressure becomes smaller than the torque of the release clutch side rotation member. That is, the start timing of the inertia phase.

  The start timing of this inertia phase is constant because the torque applied to the disengagement clutch side rotating member from a part of the power source is constant. Should be. Therefore, if the start timing shifts for some reason, this shift reflects the shift in control with high accuracy. Accordingly, the release-side pressure control learning means performs learning of the release clutch-side pressure control based on the engagement pressure adjustment history by the release-side engagement pressure gradual reduction means until the start timing of the inertia phase. Learning is possible. For this reason, highly accurate learning can be performed without causing an increase in cost, and the pressure control on the release clutch side can always be highly accurate.

  According to a second aspect of the present invention, there is provided a transmission control device that combines power from a plurality of power sources and outputs the resultant power to a rotation output shaft. A transmission control device for a hybrid power unit for a vehicle, wherein the vehicle is in a coasting state and detects whether the vehicle is in a coasting state, and a shift determination for determining whether the transmission is in a downshift. And the coasting state detecting means determine that the vehicle is in the coasting state, and the shift determining unit determines that the downshift is being performed, the output of the part of the power source is a constant torque. A constant torque control means for forming a constant torque control period to be controlled, and an engagement pressure on the release clutch side of the transmission within the constant torque control period formed by the constant torque control means. A release-side engagement pressure gradual decrease means for gradually decreasing the pressure in a reference pressure pattern, and an inertia phase for detecting a start timing of the inertia phase generated during the gradual decrease period in the reference pressure pattern by the release-side engagement pressure gradual decrease means Pressure control on the release clutch side based on the adjustment history of the engagement pressure by the release side engagement pressure gradual reduction means until the start timing detected by the start timing detection means and the inertia phase start timing detection means And release side pressure control learning means for executing the learning.

  When the hybrid power unit is the above-described hybrid power unit for a vehicle, the constant torque control means sets the output of a part of the power source to a constant torque when it is determined that the vehicle is in a coasting state and is in a downshift. A constant torque control period to be controlled is formed. This is because it is easy to apply a certain stable torque from a part of the power source to the disengagement clutch side rotation member of the transmission in the coasting state and during the downshift.

  In this way, the disengagement clutch side rotating member of the transmission receives a constant stable torque directly or indirectly from a part of the power source during a constant torque control period. Then, the release side engagement pressure gradual reduction means gradually reduces the engagement pressure on the release clutch side of the transmission with the reference pressure pattern within this constant torque control period. As a result, the start timing of the inertia phase occurs.

  This deviation in the start timing of the inertia phase reflects the deviation in control with high accuracy as described above. Accordingly, the release-side pressure control learning means performs learning of the release clutch-side pressure control based on the engagement pressure adjustment history by the release-side engagement pressure gradual reduction means until the start timing of the inertia phase. Learning is possible. For this reason, highly accurate learning can be performed without causing an increase in cost, and the pressure control on the release clutch side can always be highly accurate.

  According to a third aspect of the present invention, the transmission control device according to the third aspect is characterized in that the learning result by the release side pressure control learning means is used at least for the pressure control in the drive running state.

Thus, by using the learning result for the pressure control on the disengagement clutch side in the drive traveling state, it is possible to perform highly accurate pressure control even in the shift during the drive traveling.
According to a fourth aspect of the present invention, in the transmission control device according to the second or third aspect, the release-side engagement pressure gradual reduction means is determined not to be in the coasting state by the coasting state detection unit, and the shift determination is performed. When it is determined by the means that the downshift is being performed, the engagement pressure on the release clutch side of the transmission is immediately reduced.

  When the learning condition is not satisfied in this way, the release side engagement pressure gradual reduction means immediately decreases the engagement pressure on the release clutch side, so that the engagement pressure is obtained when learning is not performed. Since the gradual reduction is not executed, a quick shift can be performed.

  According to a fifth aspect of the present invention, in any one of the first to fourth aspects, the plurality of power sources include an internal combustion engine and an electric motor, and the electric motor corresponds to the partial power source. It is characterized by.

As described above, the plurality of power sources can include the internal combustion engine and the electric motor, and the electric motor can be used as a part of the power source.
In a transmission control device according to a sixth aspect, in any one of the first to fifth aspects, the inertia phase start timing detecting means determines the start timing based on a rotation state of the partial power source. Features.

  Since the rotation state of some power sources changes in conjunction with the start of rotation of the disengagement clutch side rotation member, the start timing of the inertia phase can be determined from the rotation state of some power sources. Thus, the inertia phase start timing detecting means can easily detect the start timing of the inertia phase without providing a special sensor.

  According to a seventh aspect of the present invention, in the transmission control device according to any one of the first to sixth aspects, the disengagement pressure control learning unit is configured as an adjustment history of the engagement pressure by the disengagement engagement pressure gradual decrease unit. The time from the gradual decrease start to the start timing is used, and learning of the pressure control on the release clutch side is executed based on the time.

  As described above, the time from the gradual decrease start to the start timing may be used as the engagement pressure adjustment history. Since the deviation in control is reflected with high accuracy at this time, highly accurate learning is possible by executing the pressure control learning on the release clutch side based on this time.

  In the transmission control device according to claim 8, in any one of claims 1 to 6, the release-side pressure control learning unit is configured as an adjustment history of the engagement pressure by the release-side engagement pressure gradually decreasing unit. The engagement pressure instruction value at the start timing is used, and learning of pressure control on the release clutch side is executed based on the instruction value.

  Thus, the instruction value of the engagement pressure at the start timing may be used as the engagement pressure adjustment history. Since the deviation in control is reflected with high accuracy in the command value of the engagement pressure, high-accuracy learning is possible by executing the pressure control learning on the release clutch side based on the command value. .

  The transmission control apparatus according to claim 9, further comprising an engagement side pressure control learning unit that learns an engagement pressure in pressure control on the engagement clutch side of the transmission according to any one of claims 1 to 8, When learning to change the engagement pressure in the pressure control on the engagement clutch side is performed by the engagement side pressure control learning means at the same shift as the learning of the release side pressure control learning means, the release side pressure control is performed. A release-side pressure control learning invalidating means for invalidating learning by the learning means is provided.

  When learning to change the engagement pressure in the pressure control is performed by the engagement-side pressure control learning means, the pressure control on the engagement clutch side performed in the latter half of the shift is performed in the release performed in the first half. There is a possibility that pressure control on the clutch side is affected.

  Therefore, when learning to change the engagement pressure in the pressure control is performed by the engagement side pressure control learning means, even if the release side pressure control learning means has already been learned, the release side pressure control learning invalid means is By invalidating learning, it is possible to prevent a decrease in learning accuracy.

  The transmission control device according to claim 10, further comprising an engagement-side pressure control learning unit that learns an engagement pressure in pressure control on the engagement clutch side of the transmission according to any one of claims 1 to 8, When learning to reduce the engagement pressure in the pressure control on the engagement clutch side is performed by the engagement side pressure control learning unit at the same shift as the learning of the release side pressure control learning unit, the release side pressure control learning is performed. A release side pressure control learning invalidating means for invalidating learning of the means is provided.

  The influence of the engagement-side pressure control learning unit on the learning of the release-side pressure control learning unit usually occurs when the engagement-side pressure control learning unit learns to reduce the engagement pressure in the pressure control.

  Therefore, especially when the learning for reducing the engagement pressure in the pressure control is performed by the engagement-side pressure control learning means, even if the release-side pressure control learning means has already been learned, By invalidating learning, it is possible to effectively prevent a decrease in learning accuracy.

  The transmission control apparatus according to claim 11 synthesizes the powers of a plurality of power sources and outputs them to the rotation output shaft, and the power synthesis of a part of the power sources is a transmission that shifts by clutch-to-clutch processing. A transmission control device for a hybrid power unit that is configured to determine whether the transmission is shifting or not, and when the shifting determination unit determines that the shifting is in progress, An offset rotation control means for controlling the rotational speed of the partial power source by drive control of the partial power source so as to be an offset rotational speed with a certain difference from the synchronous rotational speed after shifting; The engagement pressure on the engagement clutch side of the transmission is determined as a reference pressure from the timing when the rotation speed of the part of the power source by the offset rotation control means becomes the offset rotation speed or a rotation speed in the vicinity of the offset rotation speed. Engagement-side engagement pressure gradual increase means that gradually increases by turn, and a shift at the rotational speed of the partial power source that occurs during the gradual increase period in the reference pressure pattern by the engagement-side engagement pressure gradual increase means The post-shift synchronous rotation speed arrival timing detection means for detecting the arrival timing to the subsequent synchronous rotation speed, and the engagement-side engagement pressure up to the arrival timing detected by the post-shift synchronous rotation speed arrival timing detection means And an engagement-side pressure control learning means for performing learning of pressure control on the engagement clutch side based on a history of adjustment of the engagement pressure by the gradual increase means.

  By means of offset rotation control means, the drive speed of some power sources is controlled at the time of shifting so that the rotation speeds of some of the power sources become offset rotation speeds that have a certain difference from the synchronized rotation speed after shifting. Is made.

  In this situation, if the engagement clutch is completely engaged and the inertia phase is completed, even if the rotational speed is controlled by the drive control of a part of the power source, this part is restrained by the engagement clutch. The power source converges to the synchronized rotational speed after the shift.

  Here, since the offset rotation speed is a constant difference from the synchronous rotation speed after the shift, if the engagement pressure gradually increasing in the reference pressure pattern by the engagement side engagement pressure gradually increasing means is shifted, The timing at which the power source reaches the synchronized rotational speed after shifting from the offset rotational speed is deviated.

  Therefore, if this arrival timing shifts for some reason, this shift reflects a shift in control with high accuracy. Therefore, the engagement-side pressure control learning means executes the engagement clutch-side pressure control learning based on the engagement pressure adjustment history by the engagement-side engagement pressure gradually increasing means up to the timing of reaching the synchronous rotation speed. Thus, highly accurate learning is possible. For this reason, highly accurate learning can be performed without incurring an increase in cost, and the pressure control on the engagement clutch side can always be highly accurate.

  A transmission control device according to a twelfth aspect of the present invention combines a power of a plurality of power sources and outputs the resultant power to a rotation output shaft. A transmission control device for a hybrid power unit for a vehicle, wherein the vehicle is in a coasting state and detects whether the vehicle is in a coasting state, and a shift determination for determining whether the transmission is in a downshift. And the coast running state detecting means determine that the vehicle is in a coasting driving state and the shift determining means determines that the gear is being down-shifted, the drive control of the partial power source causes the An offset rotation control means for controlling the rotation speed of a part of the power source so as to be an offset rotation speed having a certain difference from the synchronous rotation speed after the shift, and the offset rotation control means The engagement pressure on the engagement clutch side of the transmission is gradually increased in a reference pressure pattern from the timing at which the rotation speed of the partial power source becomes the offset rotation speed or a rotation speed near the offset rotation speed. Synchronous rotational speed after shifting in the rotational speed of the partial power source generated during the gradual increase period in the reference pressure pattern by the engagement-side engagement pressure gradual increase means and the engagement-side engagement pressure gradual increase means The post-shift synchronous rotation speed arrival timing detection means for detecting the arrival timing at the post-shift and the engagement side engagement pressure gradual increase means until the arrival timing detected by the post-shift synchronous rotation speed arrival timing detection means And an engagement-side pressure control learning means for performing learning of pressure control on the engagement clutch side based on an engagement pressure adjustment history.

  When the hybrid power unit is the above-described hybrid power unit for a vehicle, the offset rotation control means performs this control by driving control of a part of the power source when it is determined that the vehicle is in the coasting state and the downshift is being performed. The rotational speed of the power source of the unit is controlled to be the offset rotational speed.

  In this way, the part of the power source is driven to an offset rotational speed that has a certain difference from the synchronous rotational speed after the shift. Then, the engagement side engagement pressure gradual increase means gradually increases the engagement pressure on the engagement clutch side of the transmission in a reference pressure pattern. As a result, the timing for reaching the synchronized rotational speed after the shift occurs for the part of the power source.

  As described above, the deviation in the arrival timing reflects the deviation in control with high accuracy. Therefore, the engagement-side pressure control learning means performs high-precision learning by executing the engagement clutch-side pressure control learning based on the engagement pressure adjustment history by the engagement-side engagement pressure gradually increasing means until the arrival timing. Learning is possible. For this reason, highly accurate learning can be performed without incurring an increase in cost, and the pressure control on the engagement clutch side can always be highly accurate.

  In a transmission control device according to a thirteenth aspect, in the twelfth aspect, the learning result by the engagement side pressure control learning means is used for at least the pressure control in the drive running state.

In this way, by using the learning result for the pressure control on the engagement clutch side in the drive traveling state, the pressure control with high accuracy can be performed even in the shift during the drive traveling.
According to a fourteenth aspect of the present invention, in the transmission control device according to the twelfth or thirteenth aspect, the offset rotational speed is set to be higher than the synchronous rotational speed after shifting.

  By setting the offset rotational speed to be higher than the synchronous rotational speed after shifting in this way, the rotational speed of the part of the power source is lowered to the synchronous rotational speed by the engagement side engagement pressure gradual increase means. become.

  In the case of coast shift, the synchronous rotational speed is in a reduced state, so if the offset rotational speed is set lower than the synchronous rotational speed after the shift, depending on the responsiveness of the offset rotational speed control, even if the engagement is insufficient There is a risk of misjudging by reaching the synchronous rotation speed. However, since the offset rotational speed is set to be higher than the synchronous rotational speed after shifting, the arrival timing can be detected more clearly, and more accurate learning can be performed.

  In the transmission control device according to claim 15, in any one of claims 11 to 14, the plurality of power sources include an internal combustion engine and an electric motor, and the electric motor corresponds to the partial power source. It is characterized by.

As described above, the plurality of power sources can include the internal combustion engine and the electric motor, and the electric motor can be used as a part of the power source.
In a transmission control device according to a sixteenth aspect, in any one of the eleventh to fifteenth aspects, after the arrival timing, the offset rotation control means controls the partial power source to be the offset rotation speed. It is characterized in that it is stopped or the control is switched to the control in which the partial power source is set to the synchronous rotational speed after the shift.

Thus, unnecessary energy consumption can be prevented in the part of the power source after the arrival timing is detected.
The transmission control apparatus according to claim 17, wherein the engagement side pressure control learning unit according to any one of claims 11 to 16 is an adjustment history of the engagement pressure by the engagement side engagement pressure gradual increase unit. As described above, the time from the gradual increase start to the arrival timing is used, and learning of pressure control on the engagement clutch side is executed based on the time.

  In this way, the time from the gradual increase start to the arrival timing may be used as the adjustment history of the engagement pressure, and the control deviation is accurately reflected at this time, so the engagement clutch is based on this time. By performing the pressure control learning on the side, highly accurate learning is possible.

  19. The transmission control device according to claim 18, wherein the engagement side pressure control learning unit according to any one of claims 11 to 16 is an adjustment history of the engagement pressure by the engagement side engagement pressure gradual increase unit. As described above, the instruction value of the engagement pressure at the arrival timing is used, and learning of pressure control on the engagement clutch side is executed based on the instruction value.

  In this way, the instruction value of the engagement pressure at the arrival timing may be used as the adjustment history of the engagement pressure, and the control value is reflected in this instruction value with high accuracy. By performing the pressure control learning on the engagement clutch side, highly accurate learning is possible.

  In the transmission control device according to claim 19, the configuration of the transmission control device according to any one of claims 1 to 8, and the configuration of the transmission control device according to any one of claims 11 to 18. A transmission control device provided with learning to change the engagement pressure in the pressure control on the engagement clutch side by the engagement side pressure control learning means at the same shift as the learning of the release side pressure control learning means Is provided with release side pressure control learning invalidating means for invalidating learning of the release side pressure control learning means.

  When learning to change the engagement pressure in the pressure control is performed by the engagement side pressure control learning means, the release clutch in which the pressure control on the engagement clutch side performed in the latter half of the shift is performed in the first half is performed. The possibility of affecting the pressure control on the side.

  Therefore, when learning to change the engagement pressure in the pressure control is performed by the engagement side pressure control learning means, even if the release side pressure control learning means has already been learned, the release side pressure control learning invalid means is By invalidating learning, it is possible to prevent a decrease in learning accuracy.

  In the transmission control device according to claim 20, the configuration of the transmission control device according to any of claims 1 to 8, and the configuration of the transmission control device according to any of claims 11 to 18. The transmission control device includes a learning method for reducing the engagement pressure in the pressure control on the engagement clutch side by the engagement side pressure control learning unit at the same shift as the learning of the release side pressure control learning unit. Is provided with release side pressure control learning invalidating means for invalidating learning of the release side pressure control learning means.

  The influence of the engagement-side pressure control learning unit on the learning of the release-side pressure control learning unit usually occurs when the engagement-side pressure control learning unit learns to reduce the engagement pressure in the pressure control.

  Therefore, especially when the learning for reducing the engagement pressure in the pressure control is performed by the engagement-side pressure control learning means, even if the release-side pressure control learning means has already been learned, By invalidating learning, it is possible to effectively prevent a decrease in learning accuracy.

  The transmission control device according to claim 21, wherein the reference pressure pattern executed by the disengagement-side engagement pressure gradual reduction means is before gradual decrease of the engagement pressure according to any one of claims 1 to 10, 19, and 20. A period during which a constant hydraulic pressure is set as a standby hydraulic pressure is provided, and the release-side pressure control learning unit corrects the standby hydraulic pressure based on a learning result.

  When such a period for setting the standby hydraulic pressure is provided, the learning hydraulic pressure is reflected in the standby hydraulic pressure, whereby the engagement hydraulic pressure control at the time of release can be made highly accurate by the learning.

  A transmission control device according to a twenty-second aspect is the transmission control device according to any one of the eleventh to twentieth aspects, wherein the reference pressure pattern executed by the engagement side engagement pressure gradual increase means is a standby hydraulic pressure before the engagement pressure is gradually increased. A period during which the oil pressure is constant is provided, and the engagement-side pressure control learning unit corrects the standby oil pressure based on a learning result.

  Even when the engagement pressure is gradually increased, if there is a period for the standby oil pressure, the learning oil pressure is reflected in the standby oil pressure so that the engagement oil pressure control at the time of engagement can be performed with high accuracy. Can be.

[Embodiment 1]
FIG. 1 is a block diagram showing a schematic configuration of a hybrid power unit 2 for a vehicle. The vehicle hybrid power unit 2 is mounted on a vehicle, and the power of the power source 4 is transmitted to the rotation output shaft 6 and is transmitted from the rotation output shaft 6 to the drive wheels 10 through the differential 8 as a driving force. On the other hand, there is provided a motor generator 12 (hereinafter referred to as “MG2”) which is an assist power source capable of driving control (power running control) for outputting driving force for traveling or regenerative control for recovering energy. The MG 2 is connected to the rotation output shaft 6 via the transmission 14, and the power transmitted between the MG 2 and the rotation output shaft 6 is increased or decreased according to the gear ratio set by the transmission 14.

  The power source 4 is mainly composed of an internal combustion engine 16, a motor generator 18 (hereinafter referred to as "MG1"), and a planetary gear mechanism 20 for synthesizing or distributing torque between the internal combustion engine 16 and MG1. ing. The internal combustion engine 16 is a power device such as a gasoline engine or a diesel engine, and is configured to be able to electrically control operation states such as a throttle opening (intake amount), a fuel supply amount, and an ignition timing. The control is performed by an electronic control unit (E-ECU) 22 mainly composed of a microcomputer.

  MG1 is a synchronous motor, is configured to generate a function as a motor and a function as a generator, and is connected to a power storage device, here a battery 26, via an inverter 24. An electronic control unit (MG-ECU) 28 mainly composed of a microcomputer controls the inverter 24, whereby the torque (output torque and regenerative torque) of MG1 is set.

  Note that the above-described MG2 is connected to the battery 26 via the inverter 29. The MG-ECU 28 controls the inverter 29 to control driving (power running) and regeneration, and torque in each case.

  The planetary gear mechanism 20 is a carrier that holds a sun gear 20a, a ring gear 20b concentrically arranged with respect to the sun gear 20a, and a pinion gear meshing with the sun gear 20a and the ring gear 20b so as to rotate and revolve. This is a gear mechanism that generates a differential action using 20c as three rotating elements. A rotation output shaft (here, a crankshaft) 16a of the internal combustion engine 16 is connected to a carrier 20c via a damper 16b, whereby the carrier 20c serves as an input element. The rotation of the rotation output shaft 16a of the internal combustion engine 16 (engine speed Ne) is detected by the E-ECU 22 by the engine speed sensor 16c. Further, the output shaft rotational speed Sout of the rotational output shaft 6 is also detected by the E-ECU 22 by the output shaft rotational speed sensor 6a.

  MG1 is connected to the sun gear 20a, and the sun gear 20a is a reaction force element. As a result, the ring gear 20 b serves as an output element and is connected to the rotary output shaft 6. An alignment chart of the planetary gear mechanism 20 as the above-described torque distribution mechanism (also having a function as a torque synthesizing mechanism) is as shown in FIG. As a result, a part of the power of the internal combustion engine 16 can be distributed to the rotation output shaft 6 and the other part can be distributed to the MG1.

  The transmission 14 is configured by a set of Ravigneaux planetary gear mechanisms. That is, the first sun gear 14a and the second sun gear 14b are provided, and the short pinion 14c meshes with the first sun gear 14a, and the short pinion 14c and the second sun gear 14b have a longer pinion 14d having a longer axial length. Is engaged. The long pinion 14d meshes with the ring gear 14e disposed concentrically with the sun gears 14a and 14b. Each pinion 14c, 14d is held so as to rotate and revolve by a carrier 14f. Therefore, the first sun gear 14a and the ring gear 14e constitute a mechanism corresponding to a double pinion planetary gear mechanism together with the pinions 14c and 14d, and the second sun gear 14b and the ring gear 14e together with the long pinion 14d are single pinion planets. A mechanism corresponding to the gear mechanism is configured.

  The transmission 14 is provided with a first brake B1 that selectively engages / releases the first sun gear 14a and a second brake B2 that selectively engages / releases the ring gear 14e. These brakes B1 and B2 are configured such that their torque capacities change continuously according to the engagement pressure by hydraulic pressure, electromagnetic force or the like. In this embodiment, hydraulic pressure is used as the engagement pressure. The above-described MG2 is connected to the second sun gear 14b, and the carrier 14f is connected to the rotation output shaft 6.

  Therefore, in the transmission 14, the second sun gear 14b is an input element, and the carrier 14f is an output element. Then, the clutch-to-clutch process for releasing the second brake B2 and engaging the first brake B1 results in a high speed stage. The clutch-to-clutch process for releasing the first brake B1 and engaging the second brake B2 results in a low speed stage having a higher gear ratio than the high speed stage. The shift between the two shift speeds is executed based on the running state such as the vehicle speed and the required driving force (or accelerator opening). More specifically, the shift speed region is determined in advance as a map (shift diagram), and control is performed so as to set one of the shift speeds according to the detected driving state. This control is performed by adjusting the engagement pressure by the hydraulic pressure for each of the brakes B1 and B2 when an electronic control unit (T-ECU) 30 mainly composed of a microcomputer instructs the oil control valve 14g.

  The alignment chart of the transmission 14 is as shown in [B] of FIG. If the first sun gear 14a is not fixed and the ring gear 14e is fixed by releasing the first brake B1 (hydraulic drain) and engaging the second brake B2 (hydraulic apply), the low speed stage Low is set. As a result, the torque output from the MG 2 is amplified according to the gear ratio and added to the rotation output shaft 6.

  If the first sun gear 14a is fixed and the ring gear 14e is not fixed by engaging the first brake B1 (hydraulic application) and releasing the second brake B2 (hydraulic drain), the speed ratio is smaller than the low speed stage Low. High speed stage High is set.

  In the transient state at the time of shifting, the torque is affected by the torque capacity at each brake B1, B2 and the inertia torque accompanying the change in the rotational speed. Further, the torque applied to the rotation output shaft 6 becomes a positive torque in the driving state of the MG 2 and becomes a negative torque in the driven state (regenerative state).

  The hybrid power unit 2 described above operates the internal combustion engine 16 in the most efficient state as much as possible to reduce the amount of exhaust gas, and at the same time improve fuel efficiency, and perform energy regeneration to improve fuel efficiency. Therefore, when a large driving force is required, the MG 2 is driven in a state where the torque of the power source 4 is transmitted to the rotation output shaft 6 and the torque is combined with the rotation output shaft 6. In that case, when the vehicle speed is low, the transmission 14 is set to the low speed stage Low to increase the synthesized torque, and when the vehicle speed is increased thereafter, the transmission 14 is switched to the high speed stage High and the rotation speed of the MG 2 is increased. Reduce. If there is a driving force reduction operation or the like by the driver, the transmission 14 is switched from the high speed stage High to the low speed stage Low, and the MG2 is maintained at an appropriate rotation speed. As a result, the driving efficiency and power generation efficiency of MG2 are maintained in a good state to prevent fuel consumption from deteriorating.

  At the time of such a shift of the transmission 14, as described above, the T-ECU 30 performs drain hydraulic control for releasing one of the brakes B1 and B2 via the oil control valve 14g and engaging for the other. Apply hydraulic control.

  Here, processing performed at the time of coast down shift in the control executed by the T-ECU 30 is shown in FIGS. Among these, the drain hydraulic pressure control process for release is shown in the flowcharts of FIGS. 3 and 4, the apply hydraulic pressure control process for engagement is shown in the flowcharts of FIGS. 5 and 6, and the learning value interference between release and engagement is prevented. The learning interference prevention process to be performed is shown in the flowchart of FIG. Each process is repeatedly executed in a short time period. An example of this control is shown in the timing chart of FIG. In order to execute these control processes, the ECUs 22, 28, 30 mutually exchange control information such as detection data, calculation data, and determination data.

  The drain hydraulic pressure control process (FIGS. 3 and 4) will be described. When this process is started, it is first determined whether or not a coast downshift is being performed (S100). That is, whether or not the accelerator pedal is in the OFF state is detected by the E-ECU 22, and at the same time, whether or not the shift control by the T-ECU 30 itself is executing a downshift from the shift stage High to the shift stage Low. Determine whether.

  If either the accelerator pedal OFF state or the downshift is not satisfied or both are not satisfied (NO in S100), the initial value setting flag F1 and the learning execution prohibition flag F2 are cleared (S102), and the current cycle This process ends (FIG. 8: before t0).

  If the accelerator pedal OFF state and the downshift are satisfied simultaneously (yes in S100, FIG. 8: t0), it is next determined whether or not the state is before the inertia phase (S104). If the inertia phase has not yet started immediately after the start of shifting (yes in S104), it is determined whether or not the learning execution condition for the brake B1 constant pressure standby learning value GPB1CST is satisfied (S106). Here, the learning execution condition includes a condition in which the hydraulic control through the oil control valve 14g is normally performed and the rotation of the rotation output shaft 6 is stable. For example, the condition that the temperature of the hydraulic oil used for hydraulic control of the transmission 14 is not excessively high or low and exists in a normal temperature range, the condition that the brake pedal is not operated (brake off), etc. It is.

  If the learning condition is not satisfied (no in S106), the learning execution prohibition flag F2 is set (S108). Then, the non-learning fixed torque B is set to the MG2 torque target value t_TMG2ST (S110).

  When the learning condition is satisfied (yes in S106), the learning fixed torque A is set to the MG2 torque target value t_TMG2ST (S112). Here, learning fixed torque A> non-learning fixed torque B is set.

After the process of step S112 or after the processes of steps S108 and S110, the torque instruction value TMG2 for MG2 is calculated by equation 1 (S114).
[Formula 1] TMG2 (i) ← TMG2 (i-1) + DTSWP
Here, (i) indicates a value calculated in the current control cycle, and (i-1) indicates a value calculated in the previous control cycle. (I) and (i-1) have the same contents in other formulas and symbols. The MG2 sweep-up amount DTSWP is the MG2 torque gradually increasing amount. That is, this is a value for gradually increasing the torque of MG2, which is already controlled in the regenerative state (negative torque state) in the coasting state, to generate a positive torque in the second sun gear 14b.

  Next, it is determined whether or not the current MG2 torque instruction value TMG2 (i) obtained by the above equation 1 is equal to or greater than the MG2 torque target value t_TMG2ST set in step S110 or step S112 (S116). Here, if the MG2 torque instruction value TMG2 according to the equation 1 is gradually increased and TMG2 (i) <t_TMG2ST (no in S116), is the learning execution prohibition flag F2 set next? It is determined whether or not (S120).

  Here, when F2 = 1 (yes in S120), the learning condition is not satisfied, and therefore, a process of immediately setting the drain hydraulic pressure command value PB1 (i) to 0 (Pa) is performed (S122, The failure of the learning condition is represented by a dashed line in FIG.

  If F2 = 0 (no in S120), it is next determined whether or not an initial value setting flag F1 = 0 (S124). Since the initial value is F1 = 0 (yes in S124), it is next determined whether or not the initial drain process in the first brake B1 of the transmission 14 has been completed (S126). Since the initial drain process is not started at first (No in S126), the initial drain process is executed (S128). That is, in order to execute quick pressure reduction control of the first brake B1, a process of temporarily setting the drain hydraulic pressure instruction value PB1 (i) for the first brake B1 to 0 (Pa) with respect to the oil control valve 14g. Performed (FIG. 8: t0 to t1).

  Then, when the control cycle in which the initial drain process is completed (Yes in S126, FIG. 8: t1), the drain hydraulic pressure instruction value PB1 for the first brake B1 is then set to the constant-pressure standby initial value for the first brake B1 as shown in Expression 2. It is calculated from the PB1ST and the first brake B1 constant pressure standby learning value GPB1CST (S130).

[Formula 2] PB1 (i) ← PB1ST + GPB1CST
The constant pressure standby initial value PB1ST for the first brake B1 is a predetermined constant value. The first brake B1 constant pressure standby learning value GPB1CST is a value determined based on learning as described later.

  Next, an initial value setting flag F1 is set (S132). Since the initial value setting flag F1 is set, it is determined as no in step S124 from the next control cycle, and steps S126 to S132 are not executed.

  Next, it is determined whether or not the standby time has elapsed after the MG2 torque instruction value TMG2 (i) reaches the MG2 torque target value t_TMG2ST (S134). The execution of step S114 described above is repeated for each control cycle, whereby the MG2 torque instruction value TMG2 (i) gradually increases (t0) according to the above equation 1 and finally reaches the MG2 torque target value t_TMG2ST ( = A, t3). Therefore, in step S134, the timing (t5) when the standby time has elapsed after the MG2 torque instruction value TMG2 (i) reaches the MG2 torque target value t_TMG2ST is detected.

  While the timing (t5) has not been reached (no in S134), the processing in this cycle is terminated as it is, so the drain hydraulic pressure instruction value PB1 for the first brake B1 is set by the above-described equation 2. Maintains a constant value.

When the timing (t5) is reached (yes in S134), the value of the drain hydraulic pressure command value PB1 for the first brake B1 is gradually decreased according to Equation 3 (S136).
[Formula 3] PB1 (i) <-PB1 (i-1) -DSWPB1
Here, the sweep down amount DSWPB1 is a gradually decreasing amount of the engagement pressure of the first brake B1. That is, this is a value for gradually lowering the drain hydraulic pressure command value PB1 for the first brake B1 from the state where the drain hydraulic pressure command value PB1 is maintained constant at the value set in the expression 2 and starting the inertia phase.

Thereafter, the execution of step S136 is repeated until the start of the inertia phase is confirmed, and the drain hydraulic pressure command value PB1 for the first brake B1 gradually decreases (FIG. 8: t5).
Then, the MG2 rotational speed NMG2 starts to increase from the synchronous rotational speed before the shift (t7), and when it is determined that the inertia phase start timing is reached after a time when this becomes clear (t8), no in step S104 It comes to be judged.

  Therefore, it is determined whether or not the learning execution prohibition flag F2 is set (S138). Since F2 = 0 here (no in S138), the first brake B1 constant pressure standby learning value correction amount ΔGPB1CST is based on the adjustment history of the drain hydraulic pressure command value PB1 up to the inertia phase start timing (t8). (S140). When the learning execution prohibition flag F2 = 1 (yes in S138), the first brake B1 constant pressure standby learning value correction amount ΔGPB1CST is not calculated, that is, learning is not performed.

  This learning (calculation of the first brake B1 constant pressure standby learning value correction amount ΔGPB1CST) is calculated based on the value of PB1 (i) at the inertia phase start timing (t8). For example, in the region where the absolute value of the difference is large according to the difference between PB1 (i) and the target hydraulic pressure value at the start timing (t8), the constant pressure standby learning value correction amount ΔGPB1CST for the first brake B1 according to the difference. Is set.

  Alternatively, the absolute value of the difference is determined according to the difference between the target time and the time from the gradual decrease start timing (t5) of the drain hydraulic pressure command value PB1 (i) for the first brake B1 to the inertia phase start timing (t8). In the large region, the first brake B1 constant pressure standby learning value correction amount ΔGPB1CST is set according to the difference. Here, it is set as shown in [A] of FIG.

  By correcting the first brake B1 constant pressure standby learning value GPB1CST with the first brake B1 constant pressure standby learning value correction amount ΔGPB1CST determined in this way, the hydraulic pressure control of the first brake B1 is performed thereafter. Can be reflected. As shown in [A] of FIG. 9, the area where the absolute value of the difference is small is regarded as a dead area, the first brake B1 constant pressure standby learning value correction amount ΔGPB1CST = 0, and the first brake B1 constant pressure standby. The learning value GPB1CST is not substantially updated.

  After step S140 or after it is determined yes in step S138, the drain hydraulic pressure command value PB1 (i) is set to 0 (Pa) (S142, FIG. 8: t8), and this drain hydraulic pressure control process is terminated. (S144).

  The apply hydraulic pressure control process (FIGS. 5 and 6) will be described. When this process is started, first, it is determined whether or not a coast downshift is being performed (S200). If the coast down shift is not in progress (no in S200), the initial value setting flag Fa1 and the learning execution prohibition flag Fa2 are cleared (S202), and the processing of this cycle is terminated (FIG. 8: before t0).

  If it is during the coast down shift (yes in S200), it is determined whether or not the first fill for improving the response when the hydraulic pressure is increased is completed (S204). If the first fill has not been completed yet (No in S204), it is next determined whether or not the first fill execution timing has been reached (S206). In the present embodiment, as shown in FIG. 8, the first fill is executed from timings t1 to t4.

Prior to timing t1 (no in S206), the state of the applied hydraulic pressure instruction value PB2 (i) = 0 (Pa) is continued.
At timings t1 to t4, it is determined yes in step S206, and the first fill hydraulic pressure instruction value PB2STEP is set to the apply hydraulic pressure instruction value PB2 (i) (S208).

  After step S208, it is determined whether or not a learning condition for the brake B2 constant pressure standby learning value GPB2CST is satisfied (S224). This learning execution condition is the same as step S106 in FIG.

If the learning condition is not satisfied (no in S224), the learning execution prohibition flag Fa2 is set (S226).
After the learning execution prohibition flag Fa2 is set in step S226, or after it is determined yes in step S224, it is determined whether or not the drain hydraulic pressure control process (FIGS. 3 and 4) has ended (S228). ). That is, it is determined whether or not the drain hydraulic pressure control process (FIGS. 3 and 4) is completed by executing step S144 of the drain hydraulic pressure control process (FIGS. 3 and 4) during the current coast shift.

While the drain hydraulic pressure control process (FIGS. 3 and 4) is not completed (No in S228), the process of the current control cycle is terminated as it is.
Therefore, in the subsequent control cycle, until the end of the drain hydraulic pressure control process (FIGS. 3 and 4), the following process is performed on the side of the apply hydraulic pressure control process (FIGS. 5 and 6). That is, first, step S208 is repeated, and when the first fill is completed (yes in S204), it is determined whether or not the initial value setting flag Fa1 is in a collapsed state (Fa1 = 0) (S210).

  Initially, Fa1 = 0 (yes in S210), so the second brake B2 constant pressure standby initial value PB2ST and the second brake B2 constant pressure standby learning value GPB2CST are expressed by Equation 4 to indicate the apply hydraulic pressure for the second brake B2. A value PB2 is calculated (S212).

[Formula 4] PB2 (i) ← PB2ST + GPB2CST
The constant pressure standby initial value PB2ST for the second brake B2 is a preset constant value. The second brake B2 constant pressure standby learning value GPB2CST is a value determined based on learning as described later.

  Next, an initial value setting flag Fa1 is set (S214). By setting the initial value setting flag Fa1, it is determined as no in step S210 from the next control cycle, and steps S212 and S214 are not executed.

  Following step S214, step S224 described above is executed. If it is determined no in step S224, step S226 is executed. Then, after it is determined as yes in step S224 or after step S226, it is determined as no in step S228 that the drain hydraulic pressure control process is not ended, and the current control cycle is ended.

  In the next control cycle, since it is determined to be no in step S210, it is determined whether or not feedback rotational speed control described later is started for the rotational speed NMG2 of MG2 (S215). Since the process is not started at first (No in S215), the processes of Steps S224 to S228 described above are executed. Thus, until the drain hydraulic pressure control process is completed, it is determined as no in step S228, and the state in which the process in each control cycle is completed continues.

  When the inertia phase is started while repeating such processing, the drain hydraulic pressure control processing (FIGS. 3 and 4) is completed. Accordingly, the determination at step S228 is yes, and it is then determined whether or not the learning execution prohibition flag Fa2 is set (S230).

If Fa2 = 1 (yes in S230), then the feedback rotational speed control of MG2 is executed so that the target rotational speed obtained by Expression 5 is obtained (S232, FIG. 8: t6).
[Formula 5] Target rotational speed ← Synchronous rotational speed after shifting-α
The offset amount α is appropriately set, but for example, 100 rpm is set. Therefore, when the learning execution condition is not satisfied (Fa2 = 1), the target rotational speed is set slightly lower by the offset amount α than the synchronous rotational speed after the shift.

If Fa2 = 0 (no in S230), then the feedback rotational speed control of MG2 is executed so that the target rotational speed obtained by Expression 6 is reached (S234, FIG. 8: t8).
[Formula 6] Target rotational speed ← Synchronous rotational speed after shifting + α
Therefore, when the learning execution condition is satisfied (Fa2 = 0), the target rotational speed is set slightly higher by the offset amount α than the synchronous rotational speed after the shift.

  Thereafter, step S232 or step S234 is repeatedly executed for each control cycle. Once feedback control is started, the processing in steps S232 and S234 is performed according to the change in the synchronous rotational speed after the shift. This is a process for updating the rotation speed.

  After step S232 or step S234, it is determined whether or not the rotation speed NMG2 of MG2 is maintained at the synchronized rotation speed after the shift (S236). In other words, if the second brake B2 is completely engaged and the ring gear 14e is fixed, the MG2 rotational speed NMG2 is obtained even if the rotational speed control of the MG2 is controlled to deviate from the synchronous rotational speed after the shift by the offset amount α. This coincides with the synchronized rotational speed after the shift, and this state is maintained. Therefore, step S236 is executed to determine whether or not the inertia phase has ended.

If the MG2 rotation speed NMG2 is not maintained at the synchronized rotation speed after the shift (no in S236) while the inertia phase starts or is in the middle, the current control cycle is terminated as it is.
In the next control cycle, since the feedback rotation speed control of MG2 is started, it is determined to be yes in step S215. Therefore, it is next determined whether or not the learning execution prohibition flag Fa2 is not set (Fa2 = 0) (S216).

If Fa2 = 0 (yes in S216), it is next determined whether or not the rotational speed NMG2 of MG2 has the relationship of Equation 7 with respect to the synchronous rotational speed after the shift (S218).
[Formula 7] NMG2 ≧ synchronous rotation speed after shifting−NMG2END
Here, the synchronous rotation speed after the shift is calculated from the output shaft rotation speed Sout of the rotation output shaft 6 and the gear ratio after the shift for each control cycle. The rotational speed decrease value NMG2END is a determination auxiliary value for determining that the rotational speed NMG2 of MG2 has increased slightly before the synchronous rotational speed after the shift. Here, if the MG2 rotation speed NMG2 does not yet satisfy the above expression 7 (no in S218), the process proceeds to step S224 and the subsequent steps.

  If Fa2 = 1 (NO in S216), it is then determined whether or not MG2 rotational speed NMG2 has reached the target rotational speed (S220). If the target rotational speed has not yet been reached (No in S220), the process proceeds to step S224 and subsequent steps.

As described above, MG2 rotation speed NMG2 increases toward the target rotation speed by continuing the feedback rotation speed control for MG2.
Then, when the expression 7 is satisfied with the learning execution prohibition flag Fa2 = 0 (yes in S218), or when the MG2 rotational speed NMG2 reaches the target rotational speed with the learning execution prohibition flag Fa2 = 1 (S220). Yes), the calculation of Expression 8 is executed (S222). As a result, the applied hydraulic pressure command value PB2 is gradually increased (FIG. 8: t10 or t9).

[Formula 8] PB2 (i) ← PB2 (i-1) + DSWPB2
Here, the sweep-up amount DSWPB2 is a gradually increasing amount of the engagement pressure of the second brake B2. That is, it is a value for gradually increasing the applied hydraulic pressure command value PB2 for the second brake B2 from a state where it is kept constant and for complete engagement.

  Therefore, thereafter, the apply hydraulic pressure command value PB2 is gradually increased (FIG. 8: t10 to t9). If the learning execution prohibition flag Fa2 = 0, then the MG2 rotational speed NMG2 reaches the target rotational speed (= synchronous rotational speed after shifting + α).

  If the second brake B2 is engaged and the ring gear 14e is completely fixed by gradually increasing the apply hydraulic pressure instruction value PB2, the MG2 rotational speed NMG2 is obtained even if the target rotational speed is offset from the synchronous rotational speed after the shift. Forcibly converges to the synchronized rotational speed after shifting. When Fa2 = 1, it converges to the synchronized rotational speed after the shift at timing t12, and when Fa2 = 0, it converges to the synchronized rotational speed after the shift at timing t13.

As a result, the determination at step S236 is yes. Then, the feedback rotation speed control of MG2 is stopped (S238).
Next, it is determined whether or not the learning execution prohibition flag Fa2 = 1 (S240). If the learning execution prohibition flag Fa2 = 0 (no in S240), the second brake B2 constant pressure standby learning value correction amount ΔGPB2CST reaches the timing (t13) until the MG2 rotation speed NMG2 reaches the synchronized rotation speed after the shift. It is calculated based on the adjustment history of the apply hydraulic pressure command value PB2 (S242). When the learning execution prohibition flag Fa2 = 1 (yes in S240), the second brake B2 constant pressure standby learning value correction amount ΔGPB2CST is not calculated, that is, learning is not performed.

  This learning (calculation of the second brake B2 constant pressure standby learning value correction amount ΔGPB2CST) is based on the value of PB2 (i) at the timing (t13) when the MG2 rotation speed NMG2 reaches the synchronous rotation speed after the shift. Is calculated. For example, in a region where the absolute value of the difference is large according to the difference between PB2 (i) and the target hydraulic pressure value at the arrival timing (t13), the second brake B2 constant pressure standby learning value correction amount according to the difference ΔGPB2CST is set.

  Alternatively, a region where the absolute value of the difference is large according to the difference between the target time and the time from the gradual increase start timing (t10) of the apply hydraulic pressure command value PB2 (i) for the second brake B2 to the arrival timing (t13). The second brake B2 constant pressure standby learning value correction amount ΔGPB2CST is set according to the difference. Here, the setting is made as shown in [B] of FIG.

  The second brake B2 constant pressure standby learned value correction amount ΔGPB2CST thus determined is used to correct the second brake B2 constant pressure standby learned value GPB2CST, as will be described later, thereby performing subsequent hydraulic control of the second brake B2. Can be reflected. As shown in FIG. 9B, in the region where the absolute value of the difference is small, the second brake B2 constant pressure standby learning value correction amount ΔGPB2CST = 0, and the second brake B2 constant pressure standby learning value GPB2CST No updates are made.

If the learning execution prohibition flag Fa2 = 1 is satisfied (yes in S240), or after step S242, coast down shift end processing is performed (S244).
As a result, the second brake B2 hydraulic pressure is increased to the maximum (FIG. 8: t14), the feedback rotational speed control of the MG2 rotational speed NMG2 is stopped, and the control shifts to normal control during coasting, in this case, regenerative control. (T15 ~).

  The learning interference prevention process (FIG. 7) will be described. When this process is started, it is first determined whether or not it is immediately after completion of the coast down shift process (S300). That is, it is determined whether or not the apply hydraulic pressure control process (FIGS. 5 and 6) has just ended.

If it is not immediately after completion of the coast down shift process (no in S300), the current control cycle is terminated.
If the current control cycle is immediately after the coast down shift process is completed (yes in S300), it is determined whether or not the second brake B2 constant pressure standby learning value GPB2CST needs to be updated (S302). This is determined by the value of the second brake B2 constant pressure standby learning value correction amount ΔGPB2CST obtained in the apply hydraulic pressure control process (FIGS. 5 and 6) performed immediately before. That is, if ΔGPB2CST = 0, it is not necessary to update the second brake B2 constant pressure standby learning value GPB2CST, but if ΔGPB2CST ≠ 0, it is determined that it is necessary.

Here, if ΔGPB2CST ≠ 0 (yes in S302), the second brake B2 constant pressure standby learning value GPB2CST is updated by Equation 9 (S304).
[Formula 9] GPB2CST ← GPB2CST + ΔGPB2CST
In step S310, the first brake B1 constant pressure standby learning value correction amount ΔGPB1CST and the second brake B2 constant pressure standby learning value correction amount ΔGPB2CST are cleared (S310), and the current control cycle ends.

  Therefore, when the second brake B2 constant pressure standby learning value GPB2CST is updated, even if the first brake B1 constant pressure standby learning value correction amount ΔGPB1CST ≠ 0, the first brake B1 constant pressure standby learning value GPB1CST No updates are made.

  If it is determined that the second brake B2 constant pressure standby learning value GPB2CST need not be updated (No in S302), it is determined whether or not the first brake B1 constant pressure standby learning value GPB1CST needs to be updated. (S306). This is determined in the same manner as the constant pressure standby learning value GPB2CST for the second brake B2. That is, it is determined by the value of the first brake B1 constant pressure standby learning value correction amount ΔGPB1CST obtained in the drain hydraulic pressure control process (FIGS. 3 and 4) performed immediately before. If ΔGPB1CST = 0, it is not necessary to update the first brake B1 constant pressure standby learning value GPB1CST, but if ΔGPB1CST ≠ 0, it is determined to be necessary.

If ΔGPB1CST ≠ 0 (yes in S306), the constant brake standby learning value GPB1CST for the first brake B1 is updated by Expression 10 (S308).
[Formula 10] GPB1CST ← GPB1CST + ΔGPB1CST
In step S310, the first brake B1 constant pressure standby learning value correction amount ΔGPB1CST and the second brake B2 constant pressure standby learning value correction amount ΔGPB2CST are cleared (S310), and the current control cycle ends.

If ΔGPB1CST = 0 (no in S306), the current control cycle is terminated.
In the configuration described above, the T-ECU 30 functions as a transmission control device and corresponds to each unit in the claims. Step S100 of the drain hydraulic pressure control process (FIGS. 3 and 4) corresponds to the process as the coast running state detection means and the shift determination means, and steps S116 and S118 correspond to the process as the constant torque control means. Steps S120, S122, S134, and S136 correspond to processing as the release side engagement pressure gradual reduction means, step S104 corresponds to processing as inertia phase start timing detection means, and step S140 corresponds to processing as release side pressure control learning means. To do.

  Step S200 of the apply hydraulic pressure control process (FIGS. 5 and 6) corresponds to the process as the coast running state detection means and the shift determination means, and step S234 corresponds to the process as the offset rotation control means. Steps S218 and S222 are processed as the engagement side engagement pressure gradual increasing means, step S236 is processed as the post-shift synchronous rotation speed arrival timing detecting means, and step S242 is processed as the engagement side pressure control learning means. Equivalent to.

Step S302 of the learning interference prevention process (FIG. 7) corresponds to the process as the release side pressure control learning invalidating means.
According to the first embodiment described above, the following effects can be obtained.

  (I). When it is determined that the vehicle hybrid power unit 2 is in the coasting state and is in the downshift state (yes in S100), the output of a part of the power source, here, the MG2 is made to function as an electric motor so as to have a constant torque. Control. This forms a constant torque control period (FIG. 8: t3 to t8). During this constant torque control period, a constant and stable torque is applied from the MG2 to the disengagement clutch side rotation member of the transmission 14, here the first sun gear 14a engaged and fixed by the first brake B1.

  Then, within this constant torque control period, the engagement pressure (drain hydraulic pressure command value PB1) on the disengagement clutch side of the transmission 14 is gradually decreased by the reference pressure pattern, here, the sweep down amount DSWPB1 for each cycle (S136). As a result, the start timing of the inertia phase (FIG. 8: t8) occurs.

  The deviation in the start timing of the inertia phase reflects the deviation in control with high accuracy. Therefore, learning of pressure control on the release clutch side (S140) is executed based on the adjustment history of the engagement pressure (here, drain hydraulic pressure command value PB1) in step S136 at the start timing of the inertia phase. Accordingly, the first brake B1 constant pressure standby learning value correction amount ΔGPB1CST is obtained with high accuracy, and the first brake B1 constant pressure standby learning value GPB1CST is obtained with high accuracy (S308). High-accuracy learning is possible without increasing the cost of installing a torque sensor or the like, and the pressure control on the release clutch side can always be highly accurate.

  (B). When it is determined that the vehicle hybrid power unit 2 is in the coasting state and is in the downshift state (yes in S200), the MG2 rotational speed NMG2 is set to the offset rotational speed (synchronous rotational speed after shifting + α). Then, the drive output of MG2 is feedback controlled (S234).

  Then, the engagement pressure (apply hydraulic pressure command value PB2) of the ring gear 14e released by the engagement clutch side rotating member of the transmission 14, here the second brake B2, is set as a reference pressure pattern, here the sweep-up amount for each cycle. It gradually increases with DSWPB2 (S222). As a result, the arrival timing of MG2 to the synchronized rotational speed after the shift occurs.

  As described above, the deviation in the arrival timing reflects the deviation in control with high accuracy. Accordingly, learning of pressure control on the engagement clutch side (S242) is executed based on the adjustment history of the engagement pressure in step S222 at this arrival timing. As a result, the constant pressure standby learning value GPB2CST for the second brake B2 is obtained (S304). For this reason, highly accurate learning is possible without increasing the cost of installing a torque sensor and the like, and pressure control on the engagement clutch side can always be highly accurate.

  (C). By using the first brake B1 constant pressure standby learning value GPB1CST and the second brake B2 constant pressure standby learning value GPB2CST obtained in the above-described manner for hydraulic control during gear shifting under other conditions (for example, during driving), It is possible to perform highly accurate pressure control at the time of shifting even during traveling.

  (D). In the drain hydraulic pressure control process (FIGS. 3 and 4), if the learning condition is not satisfied (yes in S120), the engagement pressure (drain hydraulic pressure instruction value PB1) on the release clutch side is immediately reduced to zero. (S122). As a result, when the learning is not performed, the engagement pressure gradual reduction (S136) is not executed, so that a quick shift is possible.

  (E). Since the inertia phase start timing (t8) and the synchronized rotation speed arrival timing (t13) after the shift are determined from the MG2 rotation speed NMG2, no special rotation speed sensor for detecting the rotation state of the first sun gear 14a or the ring gear 14e is provided. Even so, the progress of the shift can be determined.

  (F). During learning to calculate the constant pressure standby learning value correction amount ΔGPB2CST for the second brake B2 in the apply hydraulic pressure control process (FIGS. 5 and 6), the target rotational speed as the offset rotational speed is set to the post-shifting rotational speed in the feedback rotational speed control of MG2. It is set higher than the synchronous rotation speed (S234). That is, when the second brake B2 is engaged, the MG2 rotational speed NMG2 is reduced to the synchronous rotational speed.

  In the case of coast shift, since the synchronous rotational speed is in a lowered state, if the target rotational speed is set to be lower than the synchronous rotational speed after the shift, the engagement of the second brake B2 depends on the responsiveness of the target rotational speed control. Even if the match is insufficient, there is a risk of erroneous determination by reaching the synchronous rotation speed. However, since the target rotational speed is set higher than the synchronous rotational speed after the shift, there is no possibility of erroneous determination, the arrival timing can be detected more clearly, and more accurate learning is possible.

  (G). In the apply hydraulic pressure control process (FIGS. 5 and 6), after MG2 reaches the synchronized rotational speed after the shift, the feedback control using MG2 as the offset rotational speed is stopped (S238). For this reason, unnecessary energy consumption can be prevented in MG2 after arrival timing detection.

  (H). When the effective constant pressure standby learning value correction amount ΔGPB2CST for the second brake B2 is calculated in step S242 of the apply hydraulic pressure control process (FIGS. 5 and 6), the engagement clutch side that is performed in the latter half of the shift is performed. This pressure control (S212) may accelerate the engagement with the ring gear 14e. This causes a possibility of affecting the pressure control on the release clutch side that is performed in the first half.

  Therefore, if ΔGPB2CST ≠ 0 is calculated in the apply hydraulic pressure control process (FIGS. 5 and 6) (FIG. 7: yes in S302), the drain brake control process (FIGS. 3 and 4) already uses the first brake B1. Even if the constant pressure standby learning value correction amount ΔGPB1CST ≠ 0 is calculated, the learning on the drain hydraulic pressure control processing side is invalidated. That is, the constant brake standby learning value GPB1CST for the first brake B1 is not updated according to the equation (10). This can prevent a decrease in learning accuracy.

[Other embodiments]
(A). In the embodiment, learning at the time of one brake release by the drain hydraulic pressure control process (FIGS. 3 and 4) and learning at the time of other brake engagement by the apply hydraulic pressure control process (FIGS. 5 and 6) are continuously performed. However, only one of them may be executed. That is, only learning at the time of brake release may be executed at the time of shifting, or only learning at the time of brake engagement may be executed at the time of shifting.

  (B). In the drain oil pressure control process (FIGS. 3 and 4) and the apply oil pressure control process (FIGS. 5 and 6), learning is performed at the time of the coast downshift, but if the output operation to the internal combustion engine 16 is in a stable state, The present invention can also be applied to a shift other than the coast down shift.

  (C). In the apply hydraulic pressure control process (FIGS. 5 and 6), the feedback control of the MG2 rotational speed NMG2 is stopped after the arrival timing of the synchronous rotational speed after the shift, but the target rotational speed is changed to the synchronous rotational speed after the shift. The feedback control may be continued by switching.

  (D). Although a two-stage transmission is used as the transmission 14, a transmission having three or more stages may be used, and the same learning as in the above-described embodiment can be executed in each coast downshift.

1 is a block diagram showing a schematic configuration of a vehicle hybrid power unit according to a first embodiment. FIG. The alignment chart in the planetary gear mechanism and transmission of the hybrid power unit for vehicles. The flowchart of the drain hydraulic pressure control process which T-ECU of the said hybrid power unit for vehicles performs. The flowchart of a drain hydraulic pressure control process similarly. The flowchart of an apply oil pressure control process similarly. The flowchart of an apply oil pressure control process similarly. The flowchart of a learning interference prevention process similarly. 4 is a timing chart illustrating an example of control according to the first embodiment. FIG. 6 is a map configuration explanatory diagram for calculating a constant pressure standby learning value correction amount for the first brake B1 and the second brake B2 used in the drain hydraulic pressure control process and the apply hydraulic pressure control process.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 2 ... Hybrid power unit, 4 ... Power source, 6 ... Rotation output shaft, 6a ... Output shaft rotation speed sensor, 8 ... Differential, 10 ... Drive wheel, 12 ... Motor generator, 14 ... Transmission, 14a ... 1st sun gear, 14b ... second sun gear, 14c ... short pinion, 14d ... long pinion, 14e ... ring gear, 14f ... carrier, 14g ... oil control valve, 16 ... internal combustion engine, 16a ... rotary output shaft (crankshaft), 16b ... damper, 16c ... engine speed sensor, 20 ... planetary gear mechanism, 20a ... sun gear, 20b ... ring gear, 20c ... carrier, 22 ... E-ECU, 24 ... inverter, 26 ... battery, 28 ... MG-ECU, 29 ... inverter, 30 ... T-ECU, B1... First brake, B2.

Claims (22)

The power of a plurality of power sources is synthesized and output to the rotary output shaft, and the power of some of the power sources is synthesized by a transmission control device in a hybrid power unit that performs transmission via a gear-to-clutch process. There,
Shift determination means for determining whether or not the transmission is shifting;
A constant torque control means for forming a constant torque control period for controlling the output of the part of the power source to a constant torque when the shift determination means determines that a shift is in progress;
Release side engagement pressure gradual reduction means for gradually reducing the engagement pressure on the release clutch side of the transmission with a reference pressure pattern within a constant torque control period formed by the constant torque control means;
Inertia phase start timing detection means for detecting the start timing of the inertia phase generated during the gradual decrease period in the reference pressure pattern by the release side engagement pressure gradual decrease means;
Release for performing pressure control learning on the release clutch side based on the adjustment history of the engagement pressure by the release side engagement pressure gradual reduction means up to the start timing detected by the inertia phase start timing detection means Side pressure control learning means;
A transmission control device comprising: Transmission control in a hybrid power unit for a vehicle in which the power of a plurality of power sources is combined and output to a rotation output shaft, and the power of some power sources is combined via a transmission that changes speed by clutch-to-clutch processing A device,
Coast running state detecting means for detecting whether or not the vehicle is in a coast running state;
Shift determining means for determining whether or not the transmission is in a downshift;
When it is determined that the coast driving state is detected by the coast driving state detecting unit and the downshift is determined by the shift determining unit, the output of the part of the power source is controlled to a constant torque. Constant torque control means for forming a constant torque control period;
Release side engagement pressure gradual reduction means for gradually reducing the engagement pressure on the release clutch side of the transmission with a reference pressure pattern within a constant torque control period formed by the constant torque control means;
Inertia phase start timing detection means for detecting the start timing of the inertia phase generated during the gradual decrease period in the reference pressure pattern by the release side engagement pressure gradual decrease means;
Release for performing pressure control learning on the release clutch side based on the adjustment history of the engagement pressure by the release side engagement pressure gradual reduction means up to the start timing detected by the inertia phase start timing detection means Side pressure control learning means;
A transmission control device comprising: 3. The transmission control apparatus according to claim 2, wherein a learning result by the release side pressure control learning means is used for at least the pressure control in a drive running state. 4. The release side engagement pressure gradual decreasing means is determined not to be in a coast driving state by the coast driving state detecting means, and is determined to be a downshift time by the shift determining means. In this case, the transmission control device immediately decreases the engagement pressure on the release clutch side of the transmission. 5. The transmission control device according to claim 1, wherein the plurality of power sources include an internal combustion engine and an electric motor, and the electric motor corresponds to the partial power source. 6. The transmission control device according to claim 1, wherein the inertia phase start timing detection means determines the start timing based on a rotation state of the partial power source. 7. The release side pressure control learning unit according to claim 1, wherein a time from the start of gradual decrease to the start timing is used as an adjustment history of the engagement pressure by the release side engagement pressure gradual decrease unit. The transmission control device performs learning of pressure control on the release clutch side based on the time. 7. The engagement pressure instruction value at the start timing according to claim 1, wherein the release side pressure control learning means uses the engagement pressure gradually decreasing means as the engagement pressure adjustment history. , And learning of pressure control on the release clutch side based on the indicated value. In any one of Claims 1-8, The engagement side pressure control learning means which learns the engagement pressure in the pressure control by the side of the engagement clutch of the transmission is provided,
When learning to change the engagement pressure in the pressure control on the engagement clutch side is performed by the engagement side pressure control learning means at the same shift as the learning of the release side pressure control learning means, the release side pressure control is performed. A transmission control apparatus comprising release side pressure control learning invalidating means for invalidating learning of learning means. In any one of Claims 1-8, The engagement side pressure control learning means which learns the engagement pressure in the pressure control by the side of the engagement clutch of the transmission is provided,
When learning to reduce the engagement pressure in the pressure control on the engagement clutch side is performed by the engagement side pressure control learning unit at the same shift as the learning of the release side pressure control learning unit, the release side pressure control learning is performed. A transmission control apparatus comprising release side pressure control learning invalidating means for invalidating learning of means. The power of a plurality of power sources is synthesized and output to the rotary output shaft, and the power of some of the power sources is synthesized by a transmission control device in a hybrid power unit that performs transmission via a gear-to-clutch process. There,
Shift determination means for determining whether or not the transmission is shifting;
When it is determined by the shift determining means that a shift is in progress, the rotational speed of the partial power source is controlled by a drive control of the partial power source, and a constant difference from the synchronous rotational speed after the shift is obtained. Offset rotation control means for controlling the offset rotation speed to be provided;
The engagement pressure on the engagement clutch side of the transmission is determined as a reference pressure from the timing when the rotation speed of the part of the power source by the offset rotation control means becomes the offset rotation speed or a rotation speed in the vicinity of the offset rotation speed. Engagement side engagement pressure gradually increasing means gradually increasing in a pattern;
Synchronous rotation after shifting that detects the arrival timing of the rotational speed of the part of the power source that has occurred during the gradual increase period in the reference pressure pattern by the engagement-side engagement pressure gradually increasing means to the synchronized rotational speed after shifting. Number arrival timing detection means;
Based on the adjustment history of the engagement pressure by the engagement side engagement pressure gradual increase means until the arrival timing detected by the post-shift synchronous rotation speed arrival timing detection means, pressure control on the engagement clutch side is performed. Engagement-side pressure control learning means for performing learning;
A transmission control device comprising: Transmission control in a hybrid power unit for a vehicle in which the power of a plurality of power sources is combined and output to a rotation output shaft, and the power of some power sources is combined via a transmission that changes speed by clutch-to-clutch processing A device,
Coast running state detecting means for detecting whether or not the vehicle is in a coast running state;
Shift determining means for determining whether or not the transmission is in a downshift;
When it is determined that the coast running state is detected by the coast running state detecting means and the downshift is determined by the shift determining means, the part of the power source is controlled by driving control of the part of the power source. Offset rotation control means for controlling the rotational speed of the power source so as to be an offset rotational speed with a certain difference from the synchronous rotational speed after shifting;
The engagement pressure on the engagement clutch side of the transmission is determined as a reference pressure from the timing when the rotation speed of the part of the power source by the offset rotation control means becomes the offset rotation speed or a rotation speed in the vicinity of the offset rotation speed. Engagement side engagement pressure gradually increasing means gradually increasing in a pattern;
Synchronous rotation after shifting that detects the arrival timing of the rotational speed of the part of the power source that has occurred during the gradual increase period in the reference pressure pattern by the engagement-side engagement pressure gradually increasing means to the synchronized rotational speed after shifting. Number arrival timing detection means;
Based on the adjustment history of the engagement pressure by the engagement side engagement pressure gradual increase means until the arrival timing detected by the post-shift synchronous rotation speed arrival timing detection means, pressure control on the engagement clutch side is performed. Engagement-side pressure control learning means for performing learning;
A transmission control device comprising: 13. The transmission control device according to claim 12, wherein the learning result by the engagement side pressure control learning means is used for at least the pressure control in a driving traveling state. 14. The transmission control device according to claim 12, wherein the offset rotational speed is set to be higher than the synchronous rotational speed after shifting. 15. The transmission control apparatus according to claim 11, wherein the plurality of power sources include an internal combustion engine and an electric motor, and the electric motor corresponds to the partial power source. 16. The offset rotation control unit according to claim 11, wherein after the arrival timing, the offset rotation control unit stops the control to set the partial power source as the offset rotation speed, or the partial power source is turned off. A transmission control device, wherein the control is switched to the control for the synchronous rotation speed after the shift. 17. The engagement side pressure control learning unit according to claim 11, wherein the engagement side pressure control learning unit is a time from the gradual increase start to the arrival timing as the adjustment history of the engagement pressure by the engagement side engagement pressure gradual increase unit. , And learning of pressure control on the engagement clutch side is executed based on the time. 17. The engagement side pressure control learning unit according to claim 11, wherein the engagement side pressure control learning unit is configured to adjust the engagement pressure at the arrival timing as an adjustment history of the engagement pressure by the engagement side engagement pressure gradually increasing unit. A transmission control apparatus, wherein an instruction value is used, and learning of pressure control on the engagement clutch side is executed based on the instruction value. A transmission control device comprising the configuration of the transmission control device according to any one of claims 1 to 8 and the configuration of the transmission control device according to any one of claims 11 to 18,
When learning to change the engagement pressure in the pressure control on the engagement clutch side is performed by the engagement side pressure control learning means at the same speed as the learning of the release side pressure control learning means, the release side pressure control learning is performed. A transmission control apparatus comprising release side pressure control learning invalidating means for invalidating learning of means. A transmission control device comprising the configuration of the transmission control device according to any one of claims 1 to 8 and the configuration of the transmission control device according to any one of claims 11 to 18,
When learning to reduce the engagement pressure in the pressure control on the engagement clutch side is performed by the engagement side pressure control learning unit at the same shift as the learning of the release side pressure control learning unit, the release side pressure control learning is performed. A transmission control apparatus comprising release side pressure control learning invalidating means for invalidating learning of means. 21. The period according to claim 1, wherein the reference pressure pattern executed by the disengagement-side engagement pressure gradual reduction means is provided with a period during which a constant hydraulic pressure is set as a standby hydraulic pressure before the engagement pressure is gradually reduced. The transmission control apparatus, wherein the release side pressure control learning means corrects the standby hydraulic pressure based on a learning result. 21. The period according to any one of claims 11 to 20, wherein the reference pressure pattern executed by the engagement side engagement pressure gradual increase means is set to a constant hydraulic pressure as a standby hydraulic pressure before the engagement pressure is gradually increased. The engagement side pressure control learning unit corrects the standby hydraulic pressure based on a learning result.

Images