This study represents a complete approach for lifetime modelling of lithium-ion batteries in automotive applications including the cell, battery and vehicle level as well as a method for validation.
Starting with an overview of known electrochemical degradation processes, the different requirements in lifetime modelling and consequently developed types of modelling are introduced. In order to accomplish the aim to cover the complete automotive operation range, calendar and cycling ageing matrices are designed. By measuring these conditions and analysing the test results, the main ageing factors are identified and a cell lifetime model is proposed by using optimal mathematical equations. In order to extend the prediction to battery level, a vehicle model is built up by considering thermal and electrical influences.
Thus, the impact of customer behaviours as well as environmental factors on lifetime can be evaluated in detail. Simulating numerous scenarios with varying inputs for driving cycle, charging strategy, location, driven mileage, vehicle design and operation strategy points out the significance of each operating condition regarding battery lifetime. Based on this, significant lifetime extensions are demonstrated by an optimized operation and cooling strategy. Moreover, an investigation of possible second life applications is accomplished. It is pointed out by simulation that batteries have, due to the low requests, a high potential to operate several more years after vehicle usage.
Finally, an algorithm to validate lifetime predictions is presented by determining the actual state of health directly from driving data. Therefore, a novel dual Kalman filter based on an equivalent circuit model is derived ansd successfully validated against laboratory measurements.