This content is not included in your SAE MOBILUS subscription, or you are not logged in.
A Reverse Engineering Method for Powertrain Parameters Characterization Applied to a P2 Plug-In Hybrid Electric Vehicle with Automatic Transmission
ISSN: 0148-7191, e-ISSN: 2688-3627
To be published on June 23, 2020 by SAE International in United States
Over the next decade, CO2 legislation will be more demanding and the automotive industry has seen in vehicle electrification a possible solution. This has led to an increasing need for advanced powertrain systems and systematic model-based control approaches, along with additional complexity. This represents a serious challenge for all the OEMs. This paper describes a novel reverse engineering methodology developed to estimate relevant but unknown powertrain data required for fuel consumption-oriented hybrid electric vehicle modelling. The main estimated quantities include high-voltage battery internal resistance, electric motor and transmission efficiency maps, torque converter and lock-up clutch operating maps, internal combustion engine and electric motor mass moment of inertia, and finally front/rear brake torque distribution. This activity introduces a list of limited and dedicated experimental tests, carried out both on road and on a chassis dynamometer, aiming at powertrain characterization thanks to a suitable post-processing algorithm. In this regard, the methodology was tested on a P2 architecture Diesel Plug-in HEV equipped with a 9-speed AT. voltage and current sensors are used to measure the electrical power exchanged between battery and electric motor; a torque sensor on the propeller shaft measures the total torque coming out from the automatic transmission. The hydraulic pressures in the four brake calipers are measured and CAN data is logged. The results of the testing campaign are then presented and discussed, e.g. the steady-state maps of the torque converter, with a thorough explanation of the operating strategies. In conclusion, the output of the methodology is experimentally validated by introducing the identified data into a 0D map-based vehicle model, used to compare the trends of the simulation versus the experimentally-measured signals over a standard driving cycle or a dedicated maneuver.