The growing need for wearable devices, fitness accessories, and biomedical equipment has led to the upsurge in research and development of thin, flexible battery research and development. Studying degradation of such power sources used in consumer electronics devices is essential from multiple perspectives, as it allows the manufacturer to determine device warranty, affects the user purchasing decision, and can also be used to inform the user of their device’s battery health and remaining useful life in real-time. In order to achieve these goals via empirical methods, batteries are generally subjected to accelerated life cycling tests with various operating conditions, and their degradation data gathered is then used to model their SOH degradation. However, in the real world, the charge-discharge depth and charge-discharge rates for every cycle are hardly constant and vary greatly for the same user over time and for different users who use their devices differently. The real task for such developed battery models is to estimate the SOH of batteries being used in real-world scenarios with such random variations of charge-discharge depth and C-rates. To this end, the current work conducts accelerated life cycling tests of batteries with random variation in these two parameters, individually and simultaneously. Finally, multiple iterations of the SOH estimation models have been presented with different predictor variables to minimize the model validation error.