Browse Publications Technical Papers 2019-28-2443
2019-11-21

Neural Network Based Hybridized Dynamic Models for Connected Vehicles - A Case Study on Turbocharger Position Prediction 2019-28-2443

Combustion engine driven vehicles operating in a connected and autonomous vehicle (CAVs) environment, the engine drive cycles are run in a regulated manner. This is due to synchronized movement of vehicles operating in connected environment. Hence, developing intelligent and faster control of airpath variable with smooth transient tracking, helps to achieve a synchronized drive cycle. With regards to this author discuss modeling of turbocharger. This is critical for airpath system variable calculation. Due to the hybridized nature of turbocharger models, predicting accurately the position of VTG without introduction of any sensing devices is key, as sensing device induces delay in action. Authors propose a model which improve the performance and capability of VTG position prediction. A neural network based supervised learning model is developed. This model is coupled with engine models which are in series application for performance evaluation. The model is trained and validated with field data. Based on performance analysis there is a strong correlation between the model and the data. The developed model is easily scaled up for various components and be used for virtualization of sensor in a vehicle due to improved response time. This forms an ideal basis for connected vehicles, as the control of powertrain components needs to synchronize with the commands coming from the adjacent vehicles. There is a harmonized transition of vehicle drive patterns due to V2V and V2X communication, which needs a synchronous action of powertrain components due to the communication request.

SAE MOBILUS

Subscribers can view annotate, and download all of SAE's content. Learn More »

Access SAE MOBILUS »

Members save up to 18% off list price.
Login to see discount.
Special Offer: Download multiple Technical Papers each year? TechSelect is a cost-effective subscription option to select and download 12-100 full-text Technical Papers per year. Find more information here.
X