An investigation into the PNGV battery model with the addition of a dynamic temperature range

The Partnership for Next Generation-Vehicles Hybrid Pulse Power Characterization (HPPC) battery model was used to develop a base-line estimation. Within the HPPC is a method to establish an estimated open circuit voltage (VOC) from a linear calculation of internal resistances and currents. The flexibility of the parameterization was tested over a changing temperature range. Algorithms were developed that predict the VOC and state of charge (SOC) based on the battery´s temperature, current draw, terminal voltage, and sampling time step; therefore, increasing the accuracy of the battery parameterization estimator (BPE). Three production vehicles with lithium based battery chemistries were used in the study. A least square method is used as in the PNGV battery model to determine initial parameters; an improvement was shown over this base-line by using a non-linear algorithm. The estimates from both algorithms were compared to the measured data as verification.