A Deep Learning Approach Towards Generating High-fidelity Diverse Synthetic Battery Datasets

Recent surge in the number of Electric Vehicles have created a need to develop inexpensive energy-dense Battery Storage Systems. Many countries across planet put place concrete measures reduce and subsequently limit vehicles powered by fossil fuels. Lithium-ion based batteries are presently dominating electric automotive sector. Energy research efforts also focussed on accurate computation State-of-Charge such provide reliable vehicle range estimates. Although estimation algorithms precise estimates, all techniques available literature presume availability superior quality battery datasets. In reality, gaining access proprietary usage datasets is very tough for scientists. Moreover, open lack diverse charge/discharge patterns needed build generalized models. Curating measurement data time consuming needs expensive equipment. To surmount limited scenarios, we introduce few Deep Learning-based methods synthesize high-fidelity datasets, these augmented synthetic will help researchers better models presence data. We released code dataset used present approach generate The augmentation introduced here alleviate challenges.