Robust Design Method for Automatic Calibration of Automatic Transmission Shift Control System
Recently, automatic transmissions have a problem whereby the calibration of control signals for the shift control system demands an enormous amount of time due to the increase in gear ratio. This study proposes the hierarchical and inverse directional des
- PDF / 1,747,305 Bytes
- 15 Pages / 439.37 x 666.142 pts Page_size
- 21 Downloads / 176 Views
ABSTRACT Recently, automatic transmissions have a problem whereby the calibration of control signals for the shift control system demands an enormous amount of time due to the increase in gear ratio. This study proposes the hierarchical and inverse directional design method in order to avoid a lot of trial and error in calibration. This method has two model-based processes: the first is the design of the target driving torque and the second is the control signals design based on the target. Through on-vehicle tests our proposed method has the same performance as the conventional one.
Keywords Power transmission Automatic transmission Hierarchy Inverse directional design
Robust design
1 Introduction Recently, automatic transmissions in passenger cars have provided a large number of steps (eight-speed (1) and nine-speed) and a wide ratio to achieve better drivability performance and better fuel efficiency. There is, however, a problem whereby the calibration of control signals for the shift control system demands an enormous amount of time because the increase in the number of steps results in an F2012-C04-020 R. Hibino (&) T. Miyabe M. Osawa Toyota Central R&D Labs Inc, Nagakute, Japan e-mail: [email protected] H. Otsubo Toyota Motor Corporation, Toyota, Japan
SAE-China and FISITA (eds.), Proceedings of the FISITA 2012 World Automotive Congress, Lecture Notes in Electrical Engineering 193, DOI: 10.1007/978-3-642-33744-4_38, Springer-Verlag Berlin Heidelberg 2013
433
434
R. Hibino et al.
increase in the number of shift patterns [1, 2]. Formerly, simulation-based calibration was often recommended as a solution, but was not essential, because the empirical calibration process in bench and vehicle tests was only adopted in the computer. The objective of this study is the development of a systematic calibration process that avoids a lot of trial and error.
2 Concept Proposal Figure 1 shows the problem of the empirical calibration process. When people adjust the control signals based on changes in driving force, acceleration, shift time and engine revolution speed, they encounter characteristic variations and nonlinearities which make calibrations difficult. For example, the hydraulic system has variations of response speed due to oil temperature changes, characteristic variations of control valves in the manufacturing, and clearance variations of the clutch pack, which cause shift shock in initial clutch engagement. An example of nonlinearities is a response speed change of oil pressure due to the difference between flow control mode and oil pressure control mode in the hydraulic system. Therefore the shift quality for the characteristic variations are guaranteed by the robustness of control signals which are made from past experiences. In this study the above mentioned calibration process is replaced with the next design problem. If a time series pattern of driving force which satisfies all shift requirements exists and this pattern still satisfies the requirements in the conditions w
Data Loading...
