Model Predictive and Sliding Mode Control Hybridization for Voltage and Average Current Control of Dual Active Bridge DC-DC Converter in Battery Electric Vehicles Powertrain

Abstract

Dual Active Bridge (DAB) DC-DC converter is of great interest in battery electric vehicles (BEVs) powertrain applications. However, to ensure such application requirements the converter needs the supervision of an appropriate controller. The global controller layout consists of both voltage and current control loops to ensure safe converter operation, in addition of a load current feed-forward loop to enhance the converter tolerance to load variation. Both Model Predictive Control (MPC) and Sliding Mode Control (SMC) are compared within various operation conditions. Based on mathematical model, MPC is easier to design than SMC. MPC gains have empirical expressions using the converter parameters only. SMC gains have to be calculated each time circuit parameters are changed. At the same time, this SMC complex design provides a much stronger tolerance to disturbance and model inaccuracies referred to the real system. This makes SMC the only appropriate choice for inner current loop control. On the other hand, both control approaches show satisfactory performance regulating the output voltage under different operation conditions. However, the MPC low complexity level, its predefined and generalized gains make this control mode more interesting for the outer voltage loop. These features make the hybridization of MPC and SMC a promising choice for DAB DC-DC converter BEVs powertrain application.