Hybrid vehicle power management modeling and refinement

The power management strategy in many hybrid vehicles is based on expert rules and thresholds. These rule-based strategies can result in good efficiency in term of fuel economy and emissions if their thresholds and rules are accurate. However, due to the complexity and the non-linearity of the hybrid powertrain, determining accurate thresholds and rules is neither explicit nor straightforward, and experts in most cases fail to define these thresholds and rules with enough accuracy. Based on this fact, the objective of this paper is to propose a method to improve this rule-based strategy by refining its thresholds and rules. To achieve this, we used an optimization method (dynamic programming) to calculate the optimal power management, determine the optimal control signals, and extract efficient thresholds and rules that can be used in real time. Finally, simulation results for the three strategies (optimal, simple and refined strategy) are shown and discussed.