State-Space Control of Electric Vehicle Charging

Abstract

The increasing adoption of high-voltage battery systems in electric vehicles presents challenges when interfacing with legacy charging infrastructure designed for lower battery voltage levels. This is because the charging voltage must exceed the battery voltage to drive current into the battery pack. To address this, the stator windings of the electric machine are utilised as inductive elements within a DC–DC boost converter, thereby reducing additional hardware requirements and enabling effective voltage boosting. However, employing the stator windings introduces significant control challenges due to nonlinear dynamics, rotor-angle-dependent inductance, magnetic coupling, and eddy current effects. These characteristics lead to phase coupling, parameter variation, and the risk of unintended torque generation. To overcome these issues, a state-space-based cascade control strategy is developed, comprising an inner current control loop and an outer voltage control loop, both incorporating integral action to ensure accurate tracking and elimination of steady-state error. The charging system is modelled and controlled in Continuous Conduction Mode (CCM), which represents the dominant operating regime during charging. The charging system is fundamentally control-affine with a state-dependent input matrix, which introduces nonlinear behaviour. The nonlinear system is linearised around a stable operating point using a first-order Taylor expansion and the Jacobian matrix, enabling controller synthesis. The inner-loop controller is designed using Linear Quadratic Regulator (LQR) techniques, while the outer-loop controller is tuned via pole placement. Model accuracy is validated through comparison with analytical transfer functions and independent circuit simulations, including verification of transient behaviour using resonance-based peak-to-peak analysis. The proposed control framework is evaluated under varying operating conditions, including changes in supply voltage, load, and rotor position. The results demonstrate stable and reliable voltage boosting, effective current regulation, and robustness to system nonlinearities. Furthermore, the state-space formulation provides a flexible and adaptable control structure, supporting efficient integration into evolving system designs and reducing development effort for the OEM.