Adaptive Compressing Electric Vehicle Battery Pack Measurements using Polynomial Coding

This manuscript introduces an adaptive compression technique for electric vehicle battery pack measurements predicated on polynomial coding. The proposed methodology employs a lossy algorithm for voltage compression and reconstruction. The algorithm leverages the interdependence of current, time, and voltage through an adaptive time window governed by voltage thresholds. This approach facilitates the adjustment of voltage thresholds to accommodate the high dynamics of battery systems. At a Rate of Compression (RoC) of 99.2% for the LA92 drive cycle at an ambient temperature of 25°C, the root mean square error (RMSE) and mean absolute error (MAE) are 2.4 mV and 1.6 mV, respectively. The method stands out for its simplicity and effectiveness and surpasses the accuracy of current state-of-the-art compression algorithms by a factor of 20% to 160% at comparable RoC.