并联式混合动力汽车的BP网络实时能量管理.pdfVIP

12 5 Vol112 No15 2008 9 ELECTR IC MACH INES AND CONTROL Sep. 2008 BP 吴剑, 张承慧, 崔纳新 ( , 250061) : 为了提高并联式混合动力汽车( PHEV)的燃油经济性, 提出了一种基于 P神经网络的并 联式混合动力汽车实时能量 理策略采用瞬时优化能量 理策略结合多种路况离线仿真得到能 量 理规则, 利用模糊 C-均值聚类对能量 理规则进行分类并提取部分规则作为神经网络的训练 样本训练得到的 P神经网络控制器根据车辆实时工况控制混合动力系统的转矩分配, 以实现 最优的能量分配基于ADVISOR的仿真研究表明, 与瞬时优化能量 理策略相比, 该能量 理策 略不仅能够保证车辆的燃油经济性, 而且明显提高了能量 理的实时性 : 电动汽车; 并联式混合动力汽车; 神经网络; 能量 理策略; 模糊 C-均值聚类 : U469. 72 : A : 1007- 449X( 2008) 05- 0610- 05 Rea-l time energymanagement of parallel hybrid electric vehicle based on BP neural network WU Jian, ZHANG Cheng-hui, CUINa-xin ( School of Control Science and Engineering, ShandongUniversity, Jinan 250061, China) Abstract: In order to mi prove fuel economy of parallel hybrid electric vehicle ( PHEV), a rea-l tmi e ener- gy management strategy (EMS) is proposed for PHEV based on P neural network. F irstly, the energy management rules are got by offline smi ulation using instantaneous optmi ization EMS based onmany kinds of drive cycles. Then the energymanagement rules are classified by fuzzy C-mean cluster and selected as training sample of neural network. The P neural network controller is used to control the torque distribu- tion of hybrid powertain for the sake of optmi izing energy distribution. F inally, the energy m anagement strategy is mi plemented on a PHEV prototype in ADVISOR. And the smi ulation results demonstrate that, comparedw ith instantaneous optm