Systems and methods for learning torque estimate errors and updating torque estimation models are presented. In one example, torque errors are learned during an engine shut-down, after a disconnect clutch coupled between an engine and an electric machine has been released. An updated torque estimati
Systems and methods for learning torque estimate errors and updating torque estimation models are presented. In one example, torque errors are learned during an engine shut-down, after a disconnect clutch coupled between an engine and an electric machine has been released. An updated torque estimation model is then used to control torque during subsequent engine operation to improve drive feel and vehicle performance.
대표청구항▼
1. A hybrid vehicle method, comprising: attempting to provide zero torque across a disconnect clutch upon clutch release, the clutch mechanically coupling an engine to an electric machine;via a controller, learning errors in air and fueling required to provide the zero torque based on a speed profil
1. A hybrid vehicle method, comprising: attempting to provide zero torque across a disconnect clutch upon clutch release, the clutch mechanically coupling an engine to an electric machine;via a controller, learning errors in air and fueling required to provide the zero torque based on a speed profile of the engine and/or the electric machine for a duration following the clutch release; andadjusting a subsequent clutch application based on the learned errors. 2. The method of claim 1, wherein attempting to provide zero torque across the clutch includes determining an engine torque command that provides substantially zero torque and commanding the determined torque command immediately before the clutch release. 3. The method of claim 2, wherein learning errors based on the speed profile of the engine includes, if an actual engine speed profile over the duration is higher than an expected engine speed profile, learning an error corresponding to an over-estimation of engine torque; andif the actual engine speed profile over the duration is lower than the expected engine speed profile, learning an error corresponding to an under-estimation of engine torque, wherein the expected engine speed profile is based on the engine torque command. 4. The method of claim 1, wherein the errors are learned as a function of engine speed and one or more engine actuator settings. 5. The method of claim 1, wherein learning errors includes learning errors during engine idle-stop conditions when a vehicle speed is higher than a threshold. 6. A hybrid vehicle method, comprising: generating an engine torque command via a controller that provides substantially zero torque across a disconnect clutch immediately before a clutch release, the clutch mechanically coupling an engine to an electric machine;upon clutch release, learning errors in air and fueling required to provide the zero torque based on a speed profile of the engine and the electric machine for a duration following the clutch release; andadjusting clutch application based on the learned errors on a subsequent engine restart operation. 7. The method of claim 6, wherein the clutch release includes gradually releasing the disconnect clutch, while maintaining the generated engine torque command. 8. The method of claim 6, further comprising learning torque errors based on the speed profile. 9. The method of claim 8, wherein the torque errors are learned as a function of engine settings and actuator settings. 10. The method of claim 6, further comprising updating an estimation model with the learned errors and generating subsequent engine torque commands based on the updated estimation model. 11. The method of claim 6, further comprising generating subsequent engine torque commands at a subsequent engine restart based on the learned errors. 12. The method of claim 6, wherein attempting to provide zero torque across the clutch includes determining the engine torque command that provides substantially zero torque and commanding the determined torque command immediately before the clutch release. 13. The method of claim 12, wherein learning errors based on the speed profile of the engine includes, if an actual engine speed profile over the duration is higher than an expected engine speed profile, learning an error corresponding to an over-estimation of engine torque; andif the actual engine speed profile over the duration is lower than the expected engine speed profile, learning an error corresponding to an under-estimation of engine torque, wherein the expected engine speed profile is based on the engine torque command. 14. The method of claim 6, wherein the errors are learned as a function of engine speed and one or more engine actuator settings. 15. The method of claim 6, wherein learning errors includes learning errors during engine idle-stop conditions when a vehicle speed is higher than a threshold.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (2)
Wahl Robert E. ; Stokes Richard F., Gas turbine starter assist torque control system.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.