Torque estimator-based control of gear shift for shift performance increase on automatic transmissions of heavy-duty vehicles중부하 차량용 자동 변속기의 변속 성능 향상을 위한 토크 추정 기반 변속 제어

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 203
  • Download : 0
The Automatic Transmission (AT) consists of a torque converter and complex multi-planetary gears for transmitting power from the engine to the wheels. While the torque converter operates to export high torque at low speeds, multi-planetary gears with clutch packs are controlled by automated shift strategy to minimize shift shock. While such structures have been widely adopted in passenger/commercial vehicles, it is well suited for heavy-duty vehicles, especially for tracked vehicles: main battle tanks or infantry fighting vehicles mainly operated on offroad or mountainous terrain, which requires high torque of low speeds in operating environment. Heavy-duty vehicles use pilot-actuated hydraulic systems to cover a large piston volume and clutch, which centrifugal pressure affects; It shows slow dynamic response due to higher engine and vehicle inertia. Recently, new types of transmission such as dual clutch transmission or continuously variable transmission have been introduced to catch up the vehicle development trends, which emphasizes the high fuel efficiency along with compact size. These trends also affect the heavy-duty automatic transmission market because it is developed to minimize the slip area of the torque converter. This reduces fuel consumption and eliminates the one-way clutch that occupies large volume, achieving high efficiency with compact size. Accordingly, the clutch-to-clutch shifting becomes necessary on the ATs. The shifting strategy requires precise shifting control as two clutches simultaneously go off and on in order to minimize shift shock, guaranteeing qualified shifting. Therefore, shifting control method is exceptionally crucial. Previous control methods for the clutch-to-clutch shifting are mainly based on speed target, because of the only available information due to the absence of the torque sensors. The speed-based control, however, shows its limitation of inevitably using open-loop control in filling and torque phase due to the negligible acceleration change. The open-loop control requires heavy calibration on numerous parameters based on input torque and speed information in order to avoid unfavorable engine flare or tie-up. In order to overcome limitations of open-loop control including tedious calibration, torque-based control strategy with simplicity of utilizing torque from engine to the wheel axle has been introduced. This study develops torque-based clutch control method to improve the gear shift performance (fast and smooth shift) of the automatic transmission for heavy-duty vehicles. Firstly, this study suggests the development of a torque estimator based on piston pressure, considering the absence of torque sensors. This leads to the price and volume saving issue for the mass-produced transmissions. The conventional torque estimators are mainly developed for a single gear shift using simplified single gear shift model. The proposed method, however, is suitable in estimating torque of the whole gear shift without additional signal processing based on modeling of an entire transmission. In fact, the adaptive technique is applied to compensate the variation of clutch parameters such as friction coefficient. In the next step, this study develops an identification of pressure characteristics between electric current from solenoid valve and pressure from clutch piston. It is necessary because most of the mass-produced transmissions lack piston pressure sensors while actual solenoid current can be measured from the shunt resistor connected to the solenoid valves. Previous modeling of the pilot-actuated hydraulic system cannot satisfy the real time performance due to their complexities of sub-components. The proposed method exploits a lumped modeling of the overall hydraulic system. This modeling, also, is a reduced order modeling that assures the real time performance with satisfying accuracy. These are identified based on the filling phase and non-filling phase, respectively. This is because the pressure characteristics of the filling phase is different from that of the non-filling phase, which is crucial for precise filling control. This method is implemented into real-time torque estimator based on the currents of solenoid valves. Finally, the shift controller based on the torques is developed using the above torque estimator. This study distinctly suggests the clutch control method based on the target torque during the filling phase and the torque phase. In the filling phase, target torque offset makes on-coming clutch fill intentionally. Then the appropriate clutch filling time can be pinpointed from observing change of estimated clutch torque after completing the filling. In the torque phase, the torque transfer ratio, which is very complex to obtain due to the high complexity of AT, can be easily obtained even considering the inertia effect of drive line including the engine and vehicle. In addition, right torque phase time can be found from estimated off-going torque. In the inertia phase, cooperative control based-on speed target and torque target of engine and transmission respectively, guarantees fast engagement and minimum shift shock. The proposed method works correctly not only on the brake-type clutches but also on the clutch-type clutches, which have centrifugal effects. The performance of the proposed control method is evaluated both in simulations and experiments and the results demonstrate the proposed methods can make smooth and fast shift.
Advisors
Choi, Seibumresearcher최세범researcher
Description
한국과학기술원 :기계공학과,
Publisher
한국과학기술원
Issue Date
2020
Identifier
325007
Language
eng
Description

학위논문(박사) - 한국과학기술원 : 기계공학과, 2020.2,[vii, 122 p. :]

Keywords

heavy-duty vehicle▼aautomatic transmission▼aclutch-to-clutch shift▼agear shift control▼atorque estimator; 중부하 차량▼a자동변속기▼a클러치 대 클러치 변속▼a변속 제어▼a토크 추정기

URI
http://hdl.handle.net/10203/284132
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=909341&flag=dissertation
Appears in Collection
ME-Theses_Ph.D.(박사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0