Eco-Vehicle Battery System Big-Data Analysis and Fault Mode and Fault Tree Analysis (FTA) Related Robust System Development

High-voltage battery system plays a critical role in eco-friendly vehicles due to its effect on the cost and the electric driving range of eco-friendly vehicles. In order to secure the customer pool and the competitiveness of eco-vehicle technology, vehicle electrification requires lowering the battery cost and satisfying the customer needs when driving the vehicles in the real roads, for example, maximizing powers for fun drive, increasing battery capacities for achieving appropriate trip distances, etc. Because these vehicle specifications have a critical effect on the high-voltage battery specification, the key technology of the vehicle electrification is the appropriate decision on the specification of the high-voltage battery system, such as battery capacity and power. These factors affect the size of battery system and vehicle under floor design and also the profitability of the eco-friendly vehicles. In this work, the big data of Sonata hybrid electric vehicle (HEV)/plug-in hybrid electric vehicle (PHEV) battery system has been analyzed in term of four categories: cell, thermal management, 12 volt battery, and power electronics part. Analysis results show that the ratio of charge-depleting (CD)/charge-sustaining (CS) mode driving in real roads well matches the test spec. for certifying CO₂ emission. Also, the results suggest that the reduction of 12 volt battery capacity is possible based on the distribution of customers’ parking time. We have also analyzed the temperature distribution and the driving severity using driving time, distance, and electrical load (i²). Based on the analysis results, we were able to improve the development process of the battery system specifications and also tailoring the battery system specification, which can not only satisfy the customers’ needs in the real road driving, but also reduce the cost. Furthermore, the engineering specifications can be also validated and their robustness can be achieved using the big-data analysis results.