An Integrated Experimental and Numerical Methodology for Plug-In Hybrid Electric Vehicle 0D Modelling 2019-24-0072

Governments worldwide are taking actions aiming to achieve a sustainable transportation system that can comprise of minimal pollutant and GHG emissions. Particular attention is given to the real-world emissions, i.e. to the emissions achieved in the real driving conditions, outside of a controlled testing environment. In this framework, interest in vehicle fleet electrification is rapidly growing, as it is seen as a way to simultaneously reduce pollutant and GHG emissions, while on the other hand OEMs are facing a significant increase in the number of tests which are needed to calibrate this new generation of electrified powertrains over a variety of different driving scenarios. This paper introduces an integrated and standardized methodology for hybrid electric vehicles (HEVs) testing and modelling, which reduces the testing effort in terms of time and cost, and aims to provide a “virtual test rig” on which the performance of electrified powertrains can be assessed over a wide variety of different real driving scenarios. The experimental part is based on performing customized and non-invasive powertrain instrumentation. Vehicle tear-down is avoided for components characterization, which is instead achieved via a limited set of dedicated tests: constant speed driving, accelerations, standstill starts and different type of decelerations are carried out to extract a sufficient amount of data to feed the simulation models. On the simulation side, the authors have developed and calibrated a comprehensive 0D map-based HEV model, capable of simulating various hybrid architectures with high flexibility also in real driving scenarios. This approach does not require a detailed knowledge of all the powertrain and vehicle components and can be used as a virtual test bench for preliminary assessments regarding various operating conditions and driving cycles. As test case, a P0-P2 architecture gasoline Plug-in HEV has been investigated: in this case the Energy Management System (EMS) plays a crucial role in targeting the optimal power split between different on board energy sources. Using literature data and tests under both steady-state and transient conditions, the EMS model was calibrated. As a result, energy consumption and operating modes are accurately predicted, thus demonstrating the capabilities of the proposed methodology to provide a virtual test rig which can be used in the predevelopment phase of new powertrains.