From Lithium tracing inside batteries to COVID-19 transmission tracing using Artificial Intelligence

P. Nieva
University of Waterloo,
Canada

Keywords: Li ion battery, AI, tracing

Summary:

From utilizing optical characteristics of graphite to trace lithium intercalation within a lithium-ion (Li-ion) battery to COVID-19 transmission tracing using Bluetooth signal analysis from wearable devices, the Sensors and Integrated Microsystems Laboratory (SIMSLab) has been focused on developing novel sensing methods to tackle current issues within our society. As Li-ion battery usage is increasing, the SIMSLab is developing sensors that allow for early detection of battery capacity fade and determination of state of health (SOH), that can trigger a safety protocol to avoid catastrophic failure. In this talk we will present our recent work in using embedded fiber-optic battery sensors to monitor lithium intercalation within a Li-ion battery to enhance state of charge (SOC) estimation, potentially improving the performance of battery management systems. A complementary chemically resistant embedded fiber-optic temperature sensor that tracks internal temperature changes in the Li-ion battery, crucial to both battery SOC estimation and overall safe operation will also be presented. Intelligent personal systems leverage advances in wireless, wearable and sensing technologies, coupled with artificial intelligence (AI) algorithms, in order to facilitate effective disease transmission tracking. At the start of the COVID-19 pandemic, our work pivoted towards developing a wearable solution that can reduce the potential for significant outbreaks within our communities, while maintaining the privacy of individuals. In this talk, we will also discuss advanced integrated technology-aided approaches toward the development of COVID-19 tracing methods focusing on transmission risk in different population settings. Approaches for integration of multidimensional, heterogenous data using AI will be outlined while novel transmission classification models to categorise risk will be presented.