Model Predictive Control as an Energy Management Strategy for Hybrid Electric Vehicles

2009 ◽  
Author(s):  
Balaji Sampathnarayanan ◽  
Lorenzo Serrao ◽  
Simona Onori ◽  
Giorgio Rizzoni ◽  
Steve Yurkovich
Keyword(s):  
Dynamic Programming ◽  
Electric Vehicle ◽  
Energy Management ◽  
Management Strategy ◽  
Hybrid Electric Vehicle ◽  
A Priori ◽  
Dynamic Programming Algorithm ◽  
Programming Algorithm ◽  
Energy Management Strategy ◽  
Hybrid Electric

The energy management strategy in a hybrid electric vehicle is viewed as an optimal control problem and is solved using Model Predictve Control (MPC). The method is applied to a series hybrid electric vehicle, using a linearized model in state space formulation and a linear MPC algorithm, based on quadratic programming, to find a feasible suboptimal solution. The significance of the results lies in obtaining a real-time implementable control law. The MPC algorithm is applied using a quasi-static simulator developed in the MATLAB environment. The MPC solution is compared with the dynamic programming solution (offline optimization). The dynamic programming algorithm, which requires the entire driving cycle to be known a-priori, guarantees the optimality and is used here as the benchmark solution. The effect of the parameters of the MPC (length of prediction horizon, type of prediction) is also investigated.

scholarly journals Research on energy management strategy considering battery life for plug-in hybrid electric vehicle

2018 ◽  
Vol 10 (9) ◽  
pp. 168781401879776 ◽  
Author(s):  
Jianjun Hu ◽  
Zhihua Hu ◽  
Xiyuan Niu ◽  
Qin Bai
Keyword(s):  
Fuel Consumption ◽  
Electric Vehicle ◽  
Energy Management ◽  
Management Strategy ◽  
Hybrid Electric Vehicle ◽  
Battery Life ◽  
Energy Management Strategy ◽  
Driving Cycles ◽  
Comprehensive Test ◽  
Hybrid Electric

To improve the fuel efficiency and battery life-span of plug-in hybrid electric vehicle, the energy management strategy considering battery life decay is proposed. This strategy is optimized by genetic algorithm, aiming to reduce the fuel consumption and battery life decay of plug-in hybrid electric vehicle. Besides, to acquire better drive-cycle adaptability, driving patterns are recognized with probabilistic neural network. The standard driving cycles are divided into urban congestion cycle, highway cycle, and urban suburban cycle; the optimized energy management strategies in three representative driving cycles are established; meanwhile, a comprehensive test driving cycle is constructed to verify the proposed strategies. The results show that adopting the optimized control strategies, fuel consumption, and battery’s life decay drop by 1.9% and 3.2%, respectively. While using the drive-cycle recognition, the features of different driving cycles can be identified, and based on it, the vehicle can choose appropriate control strategy in different driving conditions. In the comprehensive test driving cycle, after recognizing driving cycles, fuel consumption and battery’s life decay drop by 8.6% and 0.3%, respectively.

Sign in / Sign up

Export Citation Format

Share Document