Riassunto analitico
In response to the tightening of European regulations on carbon dioxide (CO2) and pollutant emissions, in the last years, an important evolution of powertrain layout and control strategies has taken place. New technologies have been developed to improve internal combustion engines efficiency, while more complex exhaust aftertreatment systems (EATS), including diesel oxidizing catalyst (DOC), selective catalytic reduction (SCR), lean NOx trap (LNT) and diesel particulate filter (DPF), have become mandatory to be compliant with the targets. The future regulations will set even more challenging limits, together with a more demanding test procedure based on real driving cycles (RDE), forcing internal combustion engines to be efficient and clean in almost all possible operating conditions. Therefore, optimized powertrain management will be a key point to reach this target, ensuring high conversion efficiency and reduced tailpipe emissions. Besides the use of advanced EATS, which allow to strongly reduce the final tailpipe emissions, another possibility to comply with future regulations is to act directly on pollutants formation, thus limiting engine-out emissions.
In this context, the present activity is focused on the development of two predictive emissions control functions for a gasoline lean-burn plug-in hybrid electric vehicle (PHEV) equipped with a state-of-the-art EATS. In fact, the increasing availability of data from advanced driver assistance systems (ADAS) and improved vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) connectivity allows to fully take advantage of the extreme flexibility of electric hybrid powertrains, to reduce both the fuel consumption and the pollutant emissions. In particular, the upcoming regulation will introduce a more demanding and less restrained RDE type-approval procedure, named ultra-short RDE as the minimum driving distance would be 5km. Thus, specific emissions-relevant manoeuvres (such as sudden acceleration or overtalking) would become more significant in the overall computation.
Focusing on these situations, at first an already validated vehicle and EATS models, developed in Simulink and GT-Power environments, have been integrated into a co-simulation model. Then, the emissions relevant manoeuvres have been defined, and their testing scenarios modelled by means of dSPACE software ModelDesk. Successively, two dedicated control functions have been developed and tested. The first one focuses on the specific situations where the accelerator pedal is rapidly pressed, such as when the traffic light becomes green and the vehicle starts moving. More in detail, the control strategy receives the traffic light's phase and timing information from the short-range connectivity and ADAS sensors, and then it splits the torque request between the electric motor and the engine to reduce localized emissions as much as possible. On the other hand, the second function focuses on the gear shift strategy, adapting it with respect to the scenario identified by the ADAS. In particular, if an overtake is detected, the upshift and downshift maps are modified consequently to properly select a more efficient engine operating point. Finally, both the control strategies have been tested and validated in the co-simulation environment in order to verify their effectiveness in pollutant emissions reduction.
|
