Adaptive-Efficient DP Algorithm for Hybrid Electric Vehicle Powertrain: Balancing Computational Efficiency and Accuracy

Hybrid electric vehicles (HEVs) have become a hot topic in adaptive motive and academic fields owing to their high energy efficiency and low emissions. There are a large number of candidate schemes in the configuration design and parameter optimization stages of their powertrains that require performance evaluation. To address the problem of balancing computational burden and accuracy in traditional performance simulation algorithms for HEVs, such as dynamic programming (DP), an adaptive-efficient DP algorithm is proposed in this study, which introduces optimal SOC grid density identification and an optimal control point boundary relaxation method to simplify both state variable and control variable space simultaneously. Based on the series-parallel HEV configuration, the results show that the proposed method sacrifices fuel economy by 2.03% to 2.63% compared to DP, improves the simulation calculation speed by 232–319 times under various cycle conditions, and the engine operating points of the two are also relatively similar. In addition, the fuel economy performance of this method is very close to that of the Rapid-DP algorithm, and the calculation speed is also 2–3 times that of the latter, which effectively verifies that the proposed algorithm achieves high simulation accuracy while significantly improving computational efficiency.