Power Split Strategy for a Plug-In Hybrid Electric Vehicle Using Driving Pattern Recognition and Genetic Algorithm

摘要

Hybrid and plug-in hybrid electric vehicles (HEV and PHEV) represent one of the best alternatives to conventional engine propelled vehicles since they are able to increase autonomy and reduce fuel consumption. Such vehicles require a high level control in order to establish the power split decision that minimizes fuel consumption while maintaining drivability. Since fuel consumption greatly depends on road type, this paper proposes a power split strategy which benefit from information on current and previous driving conditions. Based on the Dynamic Programming (DP) work, a chosen set of control parameters were optimized on specific driving cycles using Genetic Algorithm (GA). A driving pattern recognition module using a KNN classifier is then used to select the specific driving pattern, and consequently the relevant set of parameters, that fit the most with the current driving conditions. The efficiency of the proposed controller is demonstrated through simulations.