Electrification - Statistical and reliability aspects (NEW)

With a strong correlation between vehicle usage and charge/discharge patterns, a battery deterioration model can be found.  By combining lab test results with real vehicle usage patterns from data monitored over CAN, a reliable estimate of the useful remaining life of a battery can be determined based on how it has actually been used. Designed to analyze huge quantities of vehicle CAN data, nCode GlyphWorks offers a range of cycle-counting algorithms suitable for charge/discharge analysis. Long CAN data sequences can be characterized into small and compact datasets for use in the damage models. New algorithms can also be added using MATLAB or Python and allows for multi-variate regression analysis using industry-standard tools such as scikit-learn and PyTorch.