NASA seeks intelligent monitoring and prognostic for hybrid and/or all-electric propulsion systems. A key component of these systems is the energy storage sub-system in which Lithium-ion batteries occupy a prominent place. We propose a novel, accurate and cost-effective Multi-Domain Sensing System for Lithium-ion batteries capable of detecting thermal runways earlier and predicting Remaining Useful Life (RUL) more accurately than existing methods. This technology has the capability to detect incipient faults inside Lithium-ion cells in their early stages. This will enable the effective deployment of modern protection mechanisms that are proactive and act to isolate faults with sufficient time before catastrophic effects are detected. This capability is further exploited in our system to tackle the important problem of predicting the RUL of a Lithium-ion battery in a way that promises higher levels of accuracy. All this is accomplished within an Artificial Intelligence and Stochastic-based framework that will take Lithium-ion battery monitoring and prognostics to the next level. The early detection of faults combined with the more accurate prediction of their RUL will ensure lives and assets are protected while improving the operational and ownership cost of energy storage systems based on current or future Lithium-ion batteries.
Our technology can be used to enable proactive protection strategies to safeguard lives and assets because it can detect internal incipient Lithium-ion battery faults earlier than current methods. Thermal runway precursors can be detected on time to avoid catastrophic accidents. Our technology also enables a new level in Lithium-ion battery prognostics and condition-based monitoring making it critical in applications where Lithium-ion batteries are used, such as the NASA X-57, vertical takeoff and landing systems, and other space platforms.
Hybrid/all-electric aircrafts, renewables, electric vehicles use Lithium-ion batteries for better levels of performance. The auto industry has invested $200 billion in vehicle electrification in the next four years according to AlixPartners, and this relies on Lithium-ion batteries, which are prompt to spontaneous failures, requiring better mechanisms to detect faults and predict RUL accurately.