Development of Simulation Technologies for Efficient Parameter Design in xEV Driving Control

Takuya Morikawa･Takashi Kaminaga･Motoyuki Kimata･Yuya Nagasawa･Hirotaka Kaneko (Toyota Motor)

For improving the efficiency of parameter design in xEV driving control by utilizing simulation, elements that affect parameter design under various driving scenarios have been modeled based on physics throughout the entire vehicle system. Additionally, efficient parameter design process has been built by integrating the developed models with simulation techniques such as continuous automated execution.

HEV Transaxle Loss Estimation Using Integrated Physical and Experimental Model

Motoyuki Kimata･Takuya Morikawa･Hirotaka Kaneko (Toyota Motor)

In the parameter design of HEV driving control, it is important to consider not only the outputs of the engine and motor, rolling resistance, and aerodynamic drag, but also internal resistance, such as transaxle loss, especially under various operating conditions. To address this, an integrated loss estimation model was developed by combining a physical model, based on fundamental specifications of bearings and gears, with an experimental model that captures factors difficult to represent physically, such as oil level and lubrication conditions.

Study of Model Based Design of Engine Stop Control Parameters in xEV

Takashi Kaminaga･Takuya Morikawa･Hirotaka Kaneko (Toyota Motor)

Engine stop control is one of the most complicated parameter design items in xEV as it consists of large number of maps and constants. For simplification, reinforcement learning has been adapted with a reward function delivered from the basic functional requirements. Then, this has been translated into a simple physical formula for generalization which can be applied to vehicle development and enhance parameter design efficiency.

Development of Look-ahead Energy Management Control for Electric Commercial Vehicles with Route Information

Naohiko Matsuura･Masahiro Suzuki･Hidemasa Takayama (Hino Motors)

In this study, look-ahead energy management control using route information, such as traffic, geometry, temperature, is developed for the purpose of improving energy consumption of electric commercial vehicles. Developed controller is configured with an artificial neural network model learned with the optimization theory of dynamic programming on the condition of performing on on-board ECU, and computer simulation is conducted for heavy duty electric truck to verify the effectiveness of the controller.

Accelerating powertrain development cycles with GenAI

Jan Nowack (FEV Europe)･Jin Izawa (FEV Japan)

System development faces challenges due to the multi-dimensionality of conceptual solutions in control and design domains, particularly regarding derived use-cases. To address these challenges, FEV proposes integrating the vehicle powertrain system design and controls tool into customers' system development. This integration leverages frontloading capabilities and cost reduction potentials through early-phase design. Additionally, the AI-based vehicle powertrain system design and controls tool will accelerate efficient and faster development.