Search Results

Predicting the fuel economy capability of hybrid electric vehicle (HEV) powertrains by solving the related optimal control problem has been available for a few decades. Dynamic programming (DP) is one of the most popular techniques implemented to this end. Current research aims at integrating further powertrain modeling criteria that improve the fidelity level of the optimal HEV powertrain control behavior predicted by DP, thus corroborating the reliability of the fuel economy assessment. Dedicated methodologies need further development to avoid the curse of dimensionality which is typically associated to DP when increasing the number of control and state variables considered. This paper aims at considerably reducing the overall computational effort required by DP for HEVs by removing the state term associated to the battery state-of-charge (SOC).

Battery state-of-charge (SOC) is critical information for the vehicle energy management system and must be accurately estimated to ensure reliable and affordable electrified vehicles (xEV). However, due to the nonlinear temperature, health, and SOC dependent behaviour of Li-ion batteries, SOC estimation is still a significant automotive engineering challenge. Traditional approaches to this problem, such as electrochemical models, usually require precise parameters and knowledge from the battery composition as well as its physical response. In contrast, neural networks are a data-driven approach that requires minimal knowledge of the battery or its nonlinear behaviour. The objective of this work is to present the design process of an SOC estimator using a deep feedforward neural network (FNN) approach. The method includes a description of data acquisition, data preparation, development of an FNN, FNN tuning, and robust validation of the FNN to sensor noise.