Design and test of adaptive energy management strategy for plug-in hybrid electric vehicle considering traffic information

—Intelligence provides external information for plug-in hybrid electric vehicles (PHEVs) to optimize the energy management strategy, yet effective application of information remains challenging. To this aim, the traffic scenario model based on practical sampled traffic data is established to analyze the impacts of different traffic information. On this basis, the adaptive energy management strategy is proposed. The SOC allocation is planned using the back-propagation (BP) neural network, in which the inputs of the traffic information is obtained by analyzing correlations between different traffic indexes and the power demand, as well as their impacts on the optimal SOC planning accuracy. Ulteriorly, the adaptive equivalent consumption minimization strategy (AECMS) is proposed to track the planned SOC by calculating the engine and motor torque. The equivalence factor (EF) of AECMS is adjusted using a fuzzy controller by taking the traffic information and the planned SOC as inputs. The established traffic scenario is finally introduced into the hardware-in-the-loop (HIL) test platform, together with a vehicle controller, to evaluate the performance of the proposed strategy. Research results demonstrate that the proposed strategy only increase the fuel consumption by 2.24–4.31 % compared with global optimal results.