Regenerative Braking Algorithm for Parallel Hydraulic Hybrid Vehicles Based on Fuzzy Q-Learning

The use of regenerative braking systems is an important approach for improving the travel mileage electric vehicles, and auxiliary hydraulic energy recovery system can improve efficiency process. In this paper, we present algorithm optimizing a (HRBS) based on fuzzy Q-Learning (FQL). First, built test bench, which was used to verify accuracy simulation model. Second, combined HRBS with vehicle in ADVISOR. Third, modified control strategy by introducing FQL comparing it fuzzy-control-based strategy. results showed that power savings optimized were improved about 9.62% 8.91% after 1015 cycles under urban dynamometer driving schedule (UDDS) cycle conditions compared dynamic programming (DP) algorithm. reinforcement learning method more efficient terms than