Power Converter Modulation for Battery Soc Estimation in Electric Vehicle Powertrains: a Torque Ripple Minimization Strategy

The Battery Management System (BMS) in electric vehicles (EVs) is essential for accurate battery state of charge (SoC) estimation. Electrochemical Impedance Spectroscopy (EiS) is a proven method for this, but its traditional use has been limited to laboratories due to expensive, bulky equipment. Recent advancements, however, have made it possible to adapt EIS for on-board EV applications. This study presents a novel SoC estimation technique for lithium-ion batteries, designed for integration into EV powertrains, particularly those using Permanent Magnet Synchronous Motors (PMSM). By leveraging field-oriented control (FOC), the method injects an alternating current (AC) perturbation into the direct-axis (d-axis) stator current to generate the EIS signal. The accuracy of this technique is validated against traditional laboratory-based EIS methods. A challenge of this approach is the torque ripple it produces, which can impact vehicle performance. To address this, the paper introduces a torque ripple minimization strategy using a proportional-resonant (PR) controller. This method is particularly relevant as companies like Tesla, BMW, and Nissan increasingly use PMSM in their EVs, highlighting the broader applicability of this technique.