JP2019018603A - Hybrid vehicle control method - Google Patents

Abstract

To provide a hybrid vehicle control method that can obtain high-precision driving-force estimate values in travelling-mode switching to be made by a first clutch single operation.SOLUTION: A hybrid vehicle control method has: a function of calculating control input u by using a state feedback control method in which deviations of a plurality of state quantities and target state quantities xthereof are multiplied respectively by gain K and added up on the basis of a state equation configured so that the sum of first shaft torque corresponding to a shaft torsion angle φat a side closer to a downstream side than a first clutch C1 and motor torque Tis equal to second shaft torque T, using the plurality of state quantities; a function of setting the gain K so that the plurality of state quantities converge into the target state quantities x; and a function of calculating torque command signals to be outputted to an engine 1 and a motor 2, using the control input u.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a control method for a hybrid vehicle capable of switching between a hybrid travel mode and an electric travel mode through a first clutch engaging / disengaging operation.

Conventionally, a first clutch that connects between the upstream engine and the downstream motor generator, a second clutch that connects between the motor generator and the transmission, and the engine speed and torque command signal to the engine and motor generator. There is known a hybrid vehicle provided with a control means for outputting the signal at every cycle.
When switching from the electric travel mode to the hybrid travel mode, when the engine is requested to start, the second clutch is slipped and the first clutch is semi-engaged and cranked, and the engine speed matches the motor speed. The first and second clutches are engaged.

When the first clutch is in a semi-engaged state and an engine torque exceeding the torque capacity of the first clutch is generated, the engine speed exceeds the motor speed and the direction of action of the transmission torque of the first clutch is reversed. There was a risk of engine start shock due to sudden changes.
When the engine start request is input, the control device for the hybrid vehicle in Patent Document 1 performs cranking by slipping the second clutch while the first clutch is in a semi-engaged state, and maintains the motor speed at or above the engine speed. The motor generator is controlled to do so.
The motor rotational speed target value is obtained by adding the slip rotational speed target value of the first clutch between the input and output of the first clutch to the engine rotational speed measured value.

  The motor torque control device for a vehicle disclosed in Patent Literature 2 estimates a driving force using a disturbance observer when the electric traveling mode or the hybrid traveling mode is executed, and reduces a deviation between the target driving force and the estimated driving force value. Is provided by the feedback driving force control, and the target motor torque for obtaining the target vehicle acceleration corresponding to the acceleration at the start of the mode switching is given by the feed forward acceleration control at the time of switching the driving mode. As a result, while reducing the driving force using the disturbance observer, the shock is reduced when the traveling mode is switched.

Japanese Patent No. 5391654 Japanese Patent No. 4007347

The engagement state of the second clutch is controlled when the vehicle shift is switched.
Therefore, as in Patent Document 1, in the configuration in which the second clutch is slipped when the traveling mode is switched, it is difficult to obtain the driver requested acceleration or the energy loss during the mode switching. In addition, when the traveling mode switching timing and the shift switching timing overlap, this phenomenon becomes more prominent when the traveling mode switching is prioritized.
On the other hand, the technique of Patent Document 2 realizes the driver requested acceleration by estimating the driving force using a disturbance observer in the electric traveling mode or the hybrid traveling mode.
However, since the technique of Patent Document 2 performs the estimation of the driving force but not the shaft torsion torque, the actual number of disturbances exceeds the number of disturbances that can be estimated, and the traveling mode is switched. A deviation between the target driving force and the driving force estimated value occurs, and the optimum driving force cannot be estimated even using a disturbance observer, and feedforward acceleration control is executed.
That is, in the technique of Patent Document 2, since feedforward acceleration control is executed at the time of traveling mode switching, the torque fluctuation due to the clutch torque capacity and the surrounding elastic deformation (shaft torsion, gear, etc.) cannot be absorbed. There is a possibility that the mode switching shock cannot be sufficiently suppressed.

As a result of examination by the present inventor, when the driving mode is switched, it is possible to eliminate the decrease in the driving performance of the vehicle by switching the driving mode by the single operation of the first clutch without the operation of the second clutch. In order to obtain a high-precision driving force in the travel mode switching by the single clutch operation, it is necessary to consider the shaft twist angle upstream of the first clutch and the shaft twist angle downstream of the first clutch. I came to recognize that.
However, when the shaft torsion angles upstream and downstream of the first clutch are actually detected, the shaft torsion angle detection sensor is very expensive, and it is not easy to secure the mounting space.
It can be estimated using a disturbance observer that treats the shaft torsion angle upstream from the first clutch and the shaft torsion angle downstream from the first clutch as disturbances, but at present, fluctuations become high frequency. An effective analysis method has not been established.

  An object of the present invention is to provide a hybrid vehicle control method and the like capable of reducing a shock at the time of travel mode switching in travel mode switching by a single clutch single operation.

  The hybrid vehicle control method according to claim 1 includes: a first clutch that intermittently connects between an upstream engine and a downstream motor generator; a second clutch that intermittently connects between the motor generator and a transmission; and the engine And a control means for outputting a target rotation speed and a target torque command signal for each cycle and outputting a target rotation speed and a target torque command signal for each cycle to the motor generator, and releasing the first clutch to the motor. In a control method of a hybrid vehicle capable of switching between an electric travel mode that travels by the power of a generator and a hybrid travel mode that travels by the power of the engine and a motor generator by fastening the first clutch, at least as compared with the first clutch Upstream shaft twist angle and downstream shaft twist angle, rotation of the engine The sum of the first shaft torque corresponding to the shaft torsion angle downstream of the first clutch and the motor torque of the motor generator is calculated using a plurality of state quantities including the rotational speed and torque of the motor generator. A step of setting a target value of the state quantity in accordance with a driver operation or a vehicle state based on a state equation expressed so as to be the second shaft torque upstream of the two clutches; and the state quantity target value; Calculating a control input by state feedback control in which a gain is added to each deviation from the state quantity and adding the gain; setting the gain so that the state quantity converges to the target state quantity; and the control input; A target control command value for the engine and the motor generator is calculated using the state equation. That.

In this hybrid vehicle control method, at least the shaft torsion angles upstream and downstream of the first clutch, the engine speed, the motor generator speed and torque, and the upstream side of the second clutch. Using a plurality of state quantities including the second shaft torque, the sum of the first shaft torque and the motor torque of the motor generator corresponding to the shaft torsion angle downstream of the first clutch is upstream of the second clutch. Since the state equation is expressed so as to be the second shaft torque on the side, the shaft torsion angles and the second shaft torque on the upstream side and the downstream side of the first clutch can be taken into consideration.
A state feedback control method is used to set a target value of the state quantity in accordance with a driver operation or a vehicle state, and to add a gain to each deviation between the state quantity target value and the state quantity, Since the method includes the step of calculating the control input to the motor generator, the control input to the engine and the motor generator can be obtained in consideration of the shaft torsion angles upstream and downstream of the first clutch.
Since there is a step of setting the gain so that a plurality of state quantities converge to the target state quantity, it is possible to switch the running mode while suppressing torque fluctuations.
Since there is a step of calculating a target torque command signal to be output to the engine and the motor generator using the control input and the state equation, a control command signal capable of suppressing a travel mode switching shock is sent to the engine and the motor generator. Can be output.

According to a second aspect of the present invention, in the first aspect of the invention, the plurality of state quantities include an observable element having the engine speed and the motor generator speed and torque, and an upstream side of the first clutch. And a step of estimating the state quantity by an observer created based on the state equation, and the estimated value converges to a true value. And setting the observer gain as described above, and using the state estimation value for the unobserved element for the state feedback control.
According to this configuration, even if the shaft torsion angle and the second shaft torque on the upstream side and the downstream side of the first clutch cannot actually be observed, it is possible to estimate and travel using these estimated values. A control command signal capable of suppressing the mode switching shock can be output to the engine and the motor generator.

A third aspect of the invention is characterized in that, in the first or second aspect of the invention, an optimum regulator is used in the gain setting step, and a feedback gain is set using a weighting coefficient in the evaluation function as a parameter.
According to this configuration, the control performance can be freely set by adjusting the weighting coefficient in the evaluation function.
The invention of claim 4 is characterized in that, in the invention of claim 2 or 3, the observer gain is set by using a Kalman filter in the observer gain setting step and using a noise variance value corresponding to each of the state quantities as a parameter. .
According to this configuration, it is possible to freely set the estimation performance by adjusting the dispersion value.

The invention according to claim 5 is the invention according to claim 3 or 4, wherein the feedback gain is the second shaft torque when switching from the electric travel mode to the hybrid travel mode or when switching from the hybrid travel mode to the electric travel mode. The weighting coefficient of the engine is set to be larger than the weighting coefficients of the other state quantities, and the observer gain is set to the torque of the engine when switching from the electric driving mode to the hybrid driving mode or when switching from the hybrid driving mode to the electric driving mode. The noise variance value corresponding to is set larger than the variance values of other state quantities.
According to this configuration, it is possible to set a feedback gain that can increase the control performance, and it is possible to set an observer gain that can increase the estimation performance.

A sixth aspect of the present invention is the method according to any one of the first to fifth aspects, wherein the traveling load transmitted to the second shaft through the transmission during traveling of the vehicle is introduced as a new state quantity. The state equation with the increased number of elements is estimated using the state equation and the observer created for the increased state quantity, and the unobserved including the traveling load from the estimated value It has a function to use elements for the previous state feedback control.
According to this configuration, the state feedback control can be performed even when the traveling load transmitted to the second axis cannot be observed.

  According to the hybrid vehicle control method of the present invention, it is possible to obtain a highly accurate estimated value of the driving force in the traveling mode switching by the first clutch single operation.

1 is a schematic plan view showing a powertrain of a hybrid vehicle according to a first embodiment. It is a block diagram which shows the control system of a power train. It is a functional block diagram based on a controlled object model and a state estimation model. It is a flowchart which shows a control processing procedure. 3 is a time chart of each element according to the first embodiment. It is a time chart of each element concerning a comparative example.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
The following description is an example in which the present invention is applied to a hybrid vehicle control system, and does not limit the present invention, its application, or its use.

Embodiment 1 of the present invention will be described below with reference to FIGS.
First, the outline of the powertrain PT of the hybrid vehicle according to the first embodiment will be described.
As shown in FIG. 1, the power train PT of this hybrid vehicle includes an engine 1 as a first power source and a second power source arranged in series with the engine 1 at a downstream position of the engine 1. As a motor 2 (motor generator), an automatic transmission (hereinafter abbreviated as AT) 3 arranged in series at a downstream position of the motor 2, and a pair of left and right wheels 5. A differential gear device 4 that distributes the driving force transmitted from the output shaft 9 is provided. The engine 1 has a starter motor for starting. (Not shown)

A first clutch C <b> 1 is disposed between the engine 1 and the motor generator 2.
The first clutch C1 is constituted by a wet multi-plate clutch capable of changing the transmission torque capacity by controlling the clutch hydraulic oil flow rate and the clutch hydraulic pressure continuously or stepwise by an electromagnetic solenoid valve (not shown). The upstream end of the first clutch C1 is connected to the downstream end of the output shaft of the engine 1 via the shaft 6, and the downstream end of the first clutch C1 is connected to the output shaft of the motor 2. The upstream end and the shaft 7 are connected to each other.
A second clutch C2 is disposed between the motor 2 and AT3.
Similar to the first clutch C1, the second clutch C2 is a wet multi-plate clutch capable of changing the transmission torque capacity by controlling the clutch hydraulic oil flow rate and the clutch hydraulic pressure continuously or stepwise by an electromagnetic solenoid valve. ing. The upstream end of the second clutch C2 is connected to the downstream end of the output shaft of the motor 2 via the shaft 8, and the downstream end of the second clutch C2 is connected to the output shaft 9 of the AT3. It is connected.
The second clutch C2 only needs to be capable of intermittently transmitting the driving force between the shaft 8 and the output shaft 9, and may be installed outside the AT3 or may be installed inside the AT3. .

As shown in FIG. 1, in the powertrain PT of the hybrid vehicle, when an electric travel mode (hereinafter referred to as an EV mode) executed during low load / low speed operation or deceleration regeneration is required, the first clutch C1 is released and the second clutch C2 is engaged.
When the motor 2 is driven in this state, the output rotation of the motor 2 is transmitted to the shaft 8 on the AT3 side. The AT 3 shifts the rotational motion transmitted to the shaft 8 to the selected gear and outputs it from the output shaft 9 of the AT 3. The output from the output shaft 9 of the AT 3 reaches the left and right wheels 5 via the differential gear device 4 and travel in the EV mode is executed.
When a hybrid travel mode (hereinafter referred to as HEV mode) executed during high load / high speed operation or during power generation travel is requested, both the first and second clutches C1 and C2 are engaged.
In this state, the output rotation of the engine 1 or both the output rotation of the engine 1 and the output rotation of the motor 2 are transmitted to the shaft 8 on the AT3 side. The AT 3 shifts the rotational motion transmitted to the shaft 8 to the selected gear and outputs it from the output shaft 9 of the AT 3. The output from the output shaft 9 of the AT 3 reaches the left and right wheels 5 via the differential gear device 4 and travel in the HEV mode is executed. Note that the driving mode switching timing in the EV mode and the HEV mode is determined by a driving state (vehicle speed or the like), a driver request (acceleration or the like), a battery charge state (SOC), or the like.

As shown in FIG. 2, the power train PT is integratedly controlled by a VCM (Vehicle Control Module) 10 (control means) as an integrated controller.
The VCM 10 calculates the motor torque corresponding to the driver request while absorbing the shaft torque fluctuation input from the first clutch C1 to the shaft 7 after one cycle (one sample time), and the motor 2 based on the calculated motor torque. The target torque is created. That is, when the travel mode is switched, the VCM 10 is configured such that the sum of the shaft torque corresponding to the twist angle of the shaft 7 (first shaft torque) and the motor torque of the motor 2 becomes the shaft torque of the shaft 8 corresponding to the driver requested acceleration. A control command for the motor 2 is formed.
The VCM 10 includes a PCM 11 that outputs an operation command to an injector and a throttle (not shown) of the engine 1, an inverter 14 that controls the amount of electricity supplied to the motor 2, and electromagnetics of the first and second clutches C1 and C2. It is electrically connected to a TCM 13 that outputs an operation command signal to the solenoid valve, and periodically outputs control commands to these control modules.

  As shown in FIGS. 1 and 2, the VCM 10 includes an engine speed sensor 21, a motor speed sensor 22, a motor torque sensor 23, and an opening degree sensor that can detect the opening degree of an accelerator (not shown). 24, an SOC sensor 25 that detects the state of charge of the battery 15 that supplies electricity to the inverter 14, a vehicle speed sensor (not shown), a brake sensor (not shown), and the like. Each detection signal is inputted. As a result, the VCM 10 sets the target rotational speed and target torque of the engine 1 to the PCM 11 so as to realize the traveling state (driving state) intended by the driver according to the detected accelerator opening, vehicle speed, and the like. The target rotational speed and target torque of the motor 2 are instructed to the inverter 14.

Next, a mathematical model of the power train PT that is a control target will be described.
The following description includes a description of the control method.
First, create an engine model.
Engine model rotational speed state quantity N _e and the torque T _e, the ratio twist angle phi _e and the engine torque disturbance d of the external variable axis 6, the control input as u _e, express reference to mathematical model of the DC motor ing. Internal parameter is inertia J _e, shaft friction coefficient D _e, the torque constant K _Te, armature resistance R _e, the inductance L _e, a coefficient corresponding to the induced voltage constant K _Ee, torsional rigidity coefficient K _ce. The dot (•) represents time differentiation.
As described above, the mathematical model of the engine 1 can be expressed as the following formula (1).

Next, a motor model is created.
The motor model is expressed based on a mathematical model of a DC motor, where the state quantity is the rotational speed N _m and the torque T _m , the external variable is the specific torsion angle φ _m of the shaft 7, and the control input is u _m . The internal parameters are inertia J _m , shaft friction coefficient D _m , torque constant K _Tm , armature resistance R _m , inductance L _m , coefficient corresponding to induced voltage constant K _Tm , torsional stiffness coefficient K _cm , wheel 5 to motor 2 is a traveling load T _aT transmitted to.
As described above, the mathematical model of the motor 2 can be expressed as the following formula (2).

Next, a clutch model is created.
Since the clutch model is premised on the operation of the first clutch C1 alone without the operation of the second clutch C2 when the travel mode is changed, the specific torsion angle related to the first clutch C1 and the rotation speed of the first clutch C1 are related. Model using four equations of motion (3).
Here, the state quantity includes the clutch plate rotational speed N _ce on the engine 1 side, the specific twist angle φ _e of the shaft 6, the clutch plate rotational speed N _cm on the motor 2 side, the specific twist angle φ _m of the shaft 7, and the first clutch. C1 clutch torque transmission rate P _c and control input u _c . Further, as internal parameters, the torsional stiffness coefficients K _ce and K _cm for the engine 1 side and the motor 2 side, the torsional damping coefficients Kφ _e and Kφ _m , and the length dimensions l _e and l _{m for the} shaft 6 and the shaft 7, respectively. It is.
Note that the clutch torque transmission rate P _c of the first clutch C1 is zero in an open state, 0 <P _c <1 is in a slip state, and 1 is in an engaged state, and solenoid valve operation command values (hydraulic oil flow rate and hydraulic pressure). ), The three relations of the differential rotation of the C1 clutch and _Pc are mapped in advance by experiments or the like.

Clutch torque transmission rate P _c of the first clutch C1, since the control input u _c is controllable as a control system which is independent of the engine 1 and the motor 2, can be treated as a feed-forward term. Therefore, equation (3) becomes a linear equation of motion, and the design object becomes two variables of control inputs u _e and u _m .
Therefore, when the equation (3) is transformed using the following equation (4), the state quantities are converted into the specific torsion angle φ _{e of} the shaft 6, the specific torsion angular velocity φ _e dot of the shaft 6, and the shaft torque T _ax (the first axis). 2 shaft torque) can be transformed to formula expressed using a ratio torsional angular velocity phi _m dots axis 7. Thereby, in designing the control inputs u _e and u _m , the specific torsional angular velocities φ _e dots, φ _m dots and the shaft torque T _ax can be expressed explicitly, thereby facilitating the control.

Thus, the mathematical model of the first clutch C1 can be expressed as the following formula (5).

An equation model equation (1) of the engine 1, an equation model equation (2) of the motor 2, and an equation model equation (5) of the first clutch C1 are composed of a state quantity x (t), a control input u (t), etc. If it summarizes into one state equation, it can represent like following Formula (6).
Y is an observed value corresponding to the output.
In this embodiment, since the engine speed N _e , the motor speed N _m and the motor torque T _m are detected (observed), C is expressed as follows.

Next, the design of the control input u will be described.
In order to design the control input u that can execute the running mode switching while suppressing the torque fluctuation as a state feedback system using the state equation (6), the feedback gain K of the state feedback system is set by the optimum regulator.
First, based on the driver's request, energy management, vehicle running state, and the like, the target target state quantity x ^* , which is the state immediately after switching the running mode, is set as in the following equation (7).
For example, if the engine travels at a constant vehicle speed, N _ef = N _mf , T _ef = T _axf , T _m = 0, φ _e dot = φ _em dot = 0, φ _e = T _{ef /} K _ce .

When this target state quantity x ^* is given, the control input u is designed as in the following equation (8) using the state feedback stabilization method. Note that L in Expression (8) satisfies Conditional Expression (9).
When Expression (8) and Expression (9) are substituted into Expression (6), they can be expressed as the following Expression (10). This indicates that the state quantity x (t) converges to the target state quantity x ^* when the disturbance Δ is zero by setting the feedback gain K such that A + BK becomes a stabilization matrix.

As described above, when the travel mode is switched, in the transient response in which the state quantity x converges to the target state quantity x ^* , the first clutch C1 alone is set by setting the feedback gain K corresponding to the driver requested acceleration while suppressing the torque fluctuation. Driving mode switching shock due to operation can be reduced.
In setting the feedback gain K using the optimum regulator, an evaluation function represented by the following equation (11) is set.

Based on the equation (11), the feedback gain K can be obtained from the solution P of the Riccati algebraic equation as in the following equation (12).
Here, the weighting factors Q and R when the traveling mode is switched from the EV mode to the HEV mode are set so as to satisfy the tendency of the following equation (13). Thereby, the weight of the shaft torque T _ax of the shaft 8 is set to be the largest, and then the weight of the engine torque _Te and the motor torque T _m is set to be larger than the weights of the other state quantities.
As for the input, the weight on the engine control input u _e side is set larger than the weight on the u _m side. As a result, it is possible to realize travel mode switching control that suppresses travel mode switching shock while raising the engine torque _Te smoothly.

The weighting factors Q and R when the traveling mode is switched from the HEV mode to the EV mode are set so as to satisfy the tendency of the following equation (14). Thereby, the weight of the shaft torque T _ax of the shaft 8 is set to be the largest, and the motor torque T _m and the specific torsional angular velocity of the shaft 7 are set to be smaller in order.
As for the input, the weight on the control input u _e side of the engine 1 is set larger than the weight on the control input u _m side of the motor 2. As a result, it is possible to quickly realize travel mode switching control that suppresses travel mode switching shocks.

Next, the design of the observer will be described.
It is possible to realize a desired travel mode switching control that suppresses the travel mode switching shock by feeding back the state (6) using 8 variables as the state quantities.
However, in this embodiment, three variables of the engine speed N _e , the motor speed N _m, and the torque T _m are detected, and the other five variables are not detected (observed), so the five undetected variables are estimated. There is a need to.
Therefore, as shown in FIG. 3, an observer based on the equation (6) is designed, and five undetected variables are estimated to perform state feedback.
Note that AM is a controlled object model (= state equation) expressed for a state quantity x (t) corresponding to a variable, and EM is a state estimation model (= state equation) expressed for an estimated state quantity z (t) corresponding to a variable. = Observer).

The estimated state quantity is z (t), the estimated observation value is _yz (t), the observer gain is G, and the following equation (15) can be designed. The observer gain G is a design parameter that can adjust the estimated performance of the observer.
Next, assuming that an error between the estimated state quantity z (t) and the true value is e (t), the dynamic characteristic of the error e (t) can be expressed by the following equation (16).
This is because the estimated state quantity z (t) converges to a true value (= state quantity x (t)) when the disturbance Δ is zero by setting the observer gain G such that A + GC becomes a stabilization matrix. It is shown that.

The design problem of the observer gain G results in how to arrange the eigenvalues of A + GC. In view of this, an observer is designed using a Kalman filter in which white noise is hidden in the equation of state and using the dispersion value of the white noise as a tuning parameter. If the white noise with respect to the state quantity is expressed as δ _t and the white noise with respect to the output observation value is expressed as ε _t , the equation (6) can be grasped as the following equation (17).
Then, when white noise is represented by W and S as covariance matrices of δ _t and ε _t , it can be expressed as the following equation (18).

As described above, since the observer gain G can be designed in the same manner as the equation (12), it can be obtained from the solution P _b of the Riccati algebraic equation as in the following equation (19).
Here, the tuning parameters W and S are set so as to satisfy the tendency of the following equation (20).
Thus, it is possible to set the observer gain G estimate of the engine torque T _e is converged to fastest true value. Since the output torque and the motor torque _Tm are the same at the start of switching to switch the travel mode from the EV mode to the HEV mode, the estimation error of the output torque _Tax starts from 0, whereas the engine torque _Te The estimation error is easily affected by torque vibration and the like. Therefore, since the estimation error of the engine torque T _e is significant it affects the control performance of the output torque, as soon the engine torque estimated value is required to converge to the true value.
When the clutch torque transmission rate P _c is 0, observability is not guaranteed. Therefore, the state equations are estimated by dividing the state equations into the engine 1 and the motor 2 and designing each observer.

Next, a control processing procedure of the VCM 10 will be described based on the flowchart of FIG.
Si (i = 1, 2,...) Indicates a step for each process.
As shown in the flowchart of FIG. 4, first, in S1, information such as detection values of the sensors 21 to 25 and various maps is read, and the process proceeds to S2.
In S2, it is determined whether or not a travel mode switching command is output.
As a result of the determination in S2, when a travel mode switching command is output, the process proceeds to S3.
If the result of determination in S2 is that a travel mode switching command has not been output, processing returns.

In S3, it is determined whether or not the mode is switched from the EV mode to the HEV mode.
As a result of the determination in S3, in the case of switching from the EV mode to the HEV mode, the engine is started with the starter motor (S4).
After completing the engine start in S4, the engaging operation of the first clutch C1 is started (S5), and the process proceeds to S6.
In S6, it is determined whether or not the engagement of the first clutch C1 is completed.
As a result of the determination in S6, when the engagement of the first clutch C1 is completed, the process returns.
As a result of the determination in S6, when the engagement of the first clutch C1 is not completed, the process proceeds to S7.

In S7, it is determined whether or not the first clutch C1 is in a slip state.
Result of the determination in S7, the first clutch C1 may slip state, estimates the shaft torque corresponding to the torsional angle phi _e of the shaft 6 for each period (S8), the process proceeds to S9.
As a result of the determination in S7, when the first clutch C1 is not in the slip state, the process proceeds to S5.
In S9, after estimating the shaft torque (first shaft torque) corresponding to the twist angle phi _m of the shaft 7 in each cycle, to estimate the shaft torque corresponding to the torsional angle phi _m generated in the shaft 7 after one cycle ( The process proceeds to S10) and S11. S8 to S10 are estimated by an observer.
In S11, the output after calculating the motor torque for absorbing the variation of the shaft torque corresponding to the torsional angle phi _m input after one cycle axis 7, the motor torque control command based on the calculated motor torque T _m to the inverter 14 (S12) and return. In S11, a target control command value for the engine and motor generator after one cycle is calculated from the control target model (state equation) and the control input obtained by the state feedback control.
Thus, in practice consideration and a shaft torque corresponding to the shaft torsion angle phi _m downstream of the axial torque and the first clutch C1 in response to axial torsion angle phi _e on the upstream side of the first clutch C1 that can not be observed Feedback control can be executed, and the shock at the time of hybrid driving mode switching is reduced.

If the result of determination in S3 is not switching from the EV mode to the HEV mode, since the switching is from the HEV mode to the EV mode, the first clutch C1 is released (S13), and the process proceeds to S14.
In S14, it is determined whether or not the release of the first clutch C1 is completed.
If the release of the first clutch C1 is completed as a result of the determination in S14, the process proceeds to S15.
In S15, the engine is stopped and the process returns.
As a result of the determination in S14, when the release of the first clutch C1 is not completed, the process proceeds to S16.

In S16, it is determined whether or not the first clutch C1 is in a slip state.
Result of the determination in S16, the first clutch C1 may slip state, it estimates the shaft torque corresponding to the torsional angle phi _e of the shaft 6 for each cycle (S17), the process proceeds to S18.
As a result of the determination in S16, when the first clutch C1 is not in the slip state, the process proceeds to S13.
In S18, after estimating the shaft torque corresponding to the torsional angle of the shaft 7 in each cycle, to estimate the shaft torque corresponding to the torsional angle phi _m generated in the shaft 7 after one period (S19), the process proceeds to S20. Note that S17 to S19 are estimated by an observer.
In S20, after calculating the motor torque that absorbs the fluctuation of the shaft torque according to the torsion angle φ _m generated in the shaft 7 after one cycle, a motor torque control command is output to the inverter 14 based on the calculated motor torque (S21). ), Return. In S20, a target control command value for the engine and motor generator after one cycle is calculated from the control target model (state equation) and the control input obtained by the state feedback control.
Thus, in practice consideration and a shaft torque corresponding to the shaft torsion angle phi _m downstream of the axial torque and the first clutch C1 in response to axial torsion angle phi _e on the upstream side of the first clutch C1 that can not be observed Feedback control can be executed, and the shock at the time of switching the electric travel mode is reduced.

Next, the operation and effect of the hybrid vehicle control method will be described.
A vehicle model for state feedback control in which gains K and G are set according to the first embodiment and a comparative example model using a rotational speed control method for maintaining the motor rotational speed in a state that matches the driver request are prepared. About, the state when switching from EV mode to HEV mode was analyzed by simulation. At this time, the clutch torque transmission rate P _c of the first clutch C1 is given by feed forward in consideration of fluctuations.
As shown in FIG. 5, in the vehicle model of the first embodiment, the transmission rate of the first clutch C1 fluctuates (from hydraulic pressure and the like) from the slip start t0 of the first clutch C1 to the completion of engagement of the first clutch C1 after the engine is started. Even if the friction coefficient or the like fluctuates), the shaft 8 does not fluctuate and the switching shock is suppressed.
On the other hand, the comparative example model which calculated the case where the number of rotations of the motor was controlled (FIG. 6) is the same as that of the first clutch C1 during the period from the slip start t0 of the first clutch C1 to the completion of the engagement of the first clutch C1. The change in the transmission rate is transmitted to the shaft 8, and a switching shock is generated.
Further, when the engine torque disturbance d was vibrated in Example 1, the shaft 8 was suppressed from affecting the shaft torque. (Not shown)
Note that the engine start before the start of slip t0 is not limited to the starter motor, and the engine speed may be increased by the slip of the first clutch C1, and immediately after the engine start may be set to t0.

According to this configuration, shaft torsion angles φ _e and φ _m upstream and downstream of the first clutch C 1, the rotational speed N _e of the engine 1, the rotational speed N _m and torque T _{m of the} motor 2, and the second Since the state equation is expressed using a plurality of state quantities including the second shaft torque T _ax upstream from the clutch C2, the shaft torsion angles φ _e and φ _m can be considered as the state quantities.
Since there is a step of calculating a control input u for the engine 1 and the motor 2 using a state feedback control method of multiplying the deviations between the plurality of state quantities and these target state quantities x ^* by multiplying them by a gain K, the shaft torsion A control input u for the engine 1 and the motor 2 including the angles φ _e and φ _m as state quantities can be obtained.
Since there is a gain setting step for setting the gain K so that the plurality of state quantities converge to the target state quantity x ^*, it is possible to switch the running mode while suppressing torque fluctuation.
Since there is a torque command signal calculation step for calculating a target torque command signal to be output to the engine 1 and the motor 2 using the control input u and the state equation obtained by the gain K set in the gain setting step, the travel mode switching shock is reduced. A control command signal that can be suppressed can be output to the engine 1 and the motor 2.

The plurality of state quantities include an observable element having the rotational speed N _e of the engine 1, the rotational speed N _{m of the} motor 2 and the torque T _m , and shaft twist angles φ _e , upstream and downstream of the first clutch C 1. including an unobserved element having φ _m and the second axis torque T _ax , a step of estimating a state quantity by an observer created based on a state equation, and setting an observer gain so that the estimated value converges to a true value The control input u is obtained by using the estimated value for the unobserved element.
As a result, even if the shaft torsion angles φ _e and φ _m and the second shaft torque T _ax are not actually observable, they can be estimated, and the travel mode switching shock can be suppressed using these estimated values. A control command signal can be output to the engine 1 and the motor 2.

  In the gain setting step, the feedback gain K and the observer gain G are set as parameters for the weighting coefficient in the evaluation function of the optimum regulator and the noise variance value in the Kalman filter, respectively. Estimated performance can be set freely.

When the feedback gain K is switched from the EV mode to the HEV mode or from the HEV mode to the EV mode, the weighting coefficient of the second shaft torque _Tax is set larger than the weighting coefficients of the other state quantities, and the observer gain G is the time of switching from the switching time or the HEV mode from the EV mode to the HEV mode to the EV mode is set to be larger than the other state quantity variance of the noise of the torque T _e of the engine 1.
As a result, the feedback gain K that can increase the control performance can be set, and the observer gain G that can increase the estimation performance can be set.

Next, a modified example in which the embodiment is partially changed will be described.
1) In the above embodiment, an example has been described in which three variables of engine speed, motor speed, and torque are observed values that can be actually detected. However, if the observable condition of the following equation is satisfied, state feedback stabilization Therefore, the number of observable variables or a combination thereof can be arbitrarily set.
rank (C CA CA ² ... CA ^n-1 ) = n, n: dimension of state equation

2] In the above embodiment, an example in which the gain is calculated has been described. However, an eigenvalue may be set instead of the gain.

3) In addition, those skilled in the art can implement the present invention in a form in which various modifications are added to the above-described embodiment or a combination of the above-described embodiments without departing from the spirit of the present invention. Various modifications are also included.

1 Engine 2 Motor 10 VCM
C1 first clutch C2 second clutch

Claims (6)

A first clutch that intermittently connects between an upstream engine and a downstream motor generator; a second clutch that intermittently connects between the motor generator and a transmission; and a cycle of a target rotational speed and a target torque command signal to the engine And a control means for outputting a target rotational speed and a target torque command signal to the motor generator for each cycle, and an electric travel mode in which the first clutch is disengaged to travel with the power of the motor generator; In a hybrid vehicle control method capable of switching between a hybrid travel mode in which a first clutch is engaged and the engine and a motor generator are driven by power,
From the first clutch, a plurality of state quantities including at least an upstream shaft twist angle and a downstream shaft twist angle from the first clutch, a rotational speed of the engine, and a rotational speed and torque of the motor generator are used. Is based on a state equation expressed so that the sum of the first shaft torque corresponding to the shaft twist angle on the downstream side and the motor torque of the motor generator becomes the second shaft torque on the upstream side of the second clutch,
A control input is calculated by a step of setting a target value of the state quantity in accordance with a driver operation or a vehicle state, and a state feedback control in which a deviation is added to each of the state quantity target value and the state quantity and added. Setting the gain so that the state quantity converges to the target state quantity, calculating a target control command value for the engine and the motor generator using the control input and the state equation;
A control method for a hybrid vehicle, comprising: The plurality of state quantities have an observable element having the engine speed and the motor generator speed and torque, and an unobserved having a shaft torsion angle upstream of the first clutch and the second shaft torque. Elements and
Estimating the state quantity with an observer created based on the state equation, setting an observer gain so that the estimated value converges to a true value, and determining the state estimated value of the unobserved element as the state Steps used for feedback control;
The method for controlling a hybrid vehicle according to claim 1, further comprising:   3. The hybrid vehicle control method according to claim 1, wherein an optimum regulator is used in the gain setting step, and a feedback gain is set using a weighting coefficient in the evaluation function as a parameter.   4. The hybrid vehicle control method according to claim 2, wherein a Kalman filter is used in the observer gain setting step, and an observer gain is set using a noise variance corresponding to each state quantity as a parameter. When the feedback gain is switched from the electric travel mode to the hybrid travel mode or when the hybrid travel mode is switched to the electric travel mode, the weighting coefficient of the second shaft torque is set larger than the weighting coefficients of the other state quantities. And
When the observer gain is switched from the electric travel mode to the hybrid travel mode or when the hybrid travel mode is switched to the electric travel mode, the noise variance value corresponding to the engine torque is set to be greater than the variance values of other state quantities. The method for controlling a hybrid vehicle according to claim 3, wherein the control method is set to be large.   A traveling load transmitted to the second shaft through the transmission during vehicle traveling is introduced as a new state quantity, and the state equation and the observer created for the state quantity with an increased number of elements are used. And estimating the previous state quantity having an increased number of elements and using the unobserved elements including the running load from the estimated values for the state feedback control. The control method of the hybrid vehicle of any one of -5.