Regularized MPC for Power Management of Hybrid Energy Storage Systems with Applications in Electric Vehicles*. Issue 11 (2016)

• Record Type: Journal Article
• Authors: Amy, Théo; Kong, He; Auger, Daniel; Offer, Gregory; Longo, Stefano
• Is Part Of: IFAC-PapersOnLine. Volume 49:Issue 11(2016)
• Page Start: 265
• Page End: 270
• Publication Date: 2016
• Subjects: Model Predictive Control -- Electric Vehicles -- Power Management -- Hybrid Energy Storage Systems; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2016.08.040
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: This paper examines the application of Regularized Model Predictive Control (RMPC) for Power Management (PM) of Hybrid Energy Storage Systems (HESSs). To illustrate, we apply the idea to the PM problem of a battery-supercapacitors (SCs) powertrain to reduce battery degradation in Electric Vehicles (EVs). While the application of Quadratic MPC (QMPC) in PM of HESS is not new, the idea to examine RMPC here is motivated by its capabilities to prioritize actuator actions and efficiently allocate control effort, as advocated by recent works in the control and MPC literature. Thorough simulations have been run over standard urban test drive cycles. It is found out that QMPC and RMPC, compared to rule-based PM strategies, could reduce the battery degradation over 70%. It is also shown that RMPC can slightly outperform QMPC in reducing battery degradation. Moreover, RMPC, compared to QMPC, could potentially extend the range of that SCs can be used, thus exploiting the degree of freedom of the powertrain to a larger extent. We also make some discussions on the feasibility issues and tuning challenges that RMPC faces, among others.