Accelerated assessment of optimal fuel economy benchmarks for developing the next generation HEVs

Hybrid electric vehicles (HEVs) represent a key technology to achieve both the short- and long-term requirements imposed by CO2 emission regulations. In early stages of HEV design process, off-line HEV energy management strategies typically find implementation in order to predict the fuel economy potential of each retained powertrain architecture and set of component sizes. Dynamic Programming (DP) is usually selected at this stage as HEV off-line controller. DP can compute the global optimal solution for the HEV control problem, nevertheless it remarkably suffers from course of dimensionality, thus slowing down the entire HEV development process. To overcome this draft, near-optimal control algorithms are currently under development to quicken the prediction of fuel economy capabilities of HEV design candidates. This presentation particularly illustrates the development of a new HEV off-line energy management strategy which is proved producing comparable control results with respect to DP, while reducing the corresponding computational time by near two orders of magnitude. Different HEV architectures are considered as case studies to demonstrate the effectiveness of the algorithm including the power-split type, the parallel type and the series-parallel type. Obtained results suggest that the illustrated algorithm can potentially find implementation in HEV design methodologies to accelerate the overall vehicle development process.