Knowledge Implementation and Transfer With an Adaptive Learning Network for Real-Time Power Management of the Plug-in Hybrid Vehicle

Essential decision-making tasks such as power management in future vehicles will benefit from the development of artificial intelligence technology for safe and energy-efficient operations. To develop technique using neural network deep learning energy plug-in hybrid vehicle evaluate its advantage, this article proposes a new adaptive that incorporates deterministic policy gradient (DDPG) with an neuro-fuzzy inference system (ANFIS) network. First, ANFIS is built global K-fold fuzzy (GKFL) method real-time implementation offline dynamic programming result. Then, DDPG developed to regulate input real-world reinforcement signal. The networks are integrated maximize control utility (CU), which function vehicle’s efficiency battery state-of-charge. Experimental studies conducted testify performance robustness DDPG-ANFIS It has shown studied achieves 8% higher CU than MATLAB toolbox on vehicle. In five simulated driving conditions, increased maximum mean value by 138% over ANFIS-only 5% DDPG-only