Simulation for prediction of vehicle efficiency, performance, range and lifetime: a review of current techniques and their applicability to current and future testing standards

Computer simulation tools can give early indicators of key vehicle characteristics. In traditional hybrid vehicles, this is important in designing for optimal fuel consumption; in plug-in hybrids and pure electric vehicles, it is critical for accurate prediction of range, a key market qualifier. There are a variety of techniques, typically operating at different levels of fidelity and employing different modelling philosophies. This paper develops on earlier work, exploring conventional and `backward' techniques in the context of current NEDC-based UNECE vehicle testing standards and the proposed replacements based on the World Light Test Procedure. Model sensitivities for A, C and D-segment vehicles are quantified and this is used to explore aspects where accurate models are key and where lower-fidelity representative models are appropriate. The paper also explores the sensitivity of predictions to `PID control' driver models, and discusses the effect of cycle-following tolerance on predictions. Finally, the paper proposes new standards - suitable for simulation or real-world testing - for a common quantification of in-use battery lifetime. The use of these techniques and the sensitivity analysis methods on a representative simulation model is demonstrated as a case study, and the impacts on battery management strategy design are explored.