Data compression offers the possibility to decrease the amount of storage or the bandwidth of a communication channel required to save or transmit collected data. In battery packs of electric vehicles (EV), the data over time contains values of voltage, current, temperature, state of charge and others depending on the battery management feasibility. Compression of this battery data offers different new fields of application.
Lately, in China, it is required that all EVs transmit their battery values every 10 seconds to a cloud. With the steady rise of EV market share, this results in a large amount of data. The database offers trends in research related to cloud computation, digital twins and the field of big data analysis as well as artificial intelligence. To reduce the total bandwidth of the transmitted data, or offer the possibility to increase the sample rate below every 10 seconds, data compression is necessary.
Battery ageing algorithms track the change of battery parameters over time. To track those changes in the battery management system with adaptive algorithms, long periods of data of similar cells need to be available. The embedded systems used in battery management are limited in both computational performance and memory. Thus specific optimized data processing needs to be considered.
Those two examples show that data compression optimizes and also offers new fields of batteries in EVs. This work contains an overview of the results of testing the most common compression techniques for battery systems. The focus hereby is on lossless data compression. Lossless compression obtains all information. Thus future algorithms can use the total amount of information measured by the battery management system. The compression is applied to battery data recorded from electric vehicles.