SBIR Phase II: Forecasting Battery Health and Maintenance using Data-Driven Predictive Analytics

The broader impact of this Small Business Innovation Research (SBIR) Phase II project includes enhancing US economic competitiveness, improving the health and welfare of the American public, and developing the US technical workforce. The success of this project will have a direct impact on the manufacturers, integrators, and operators of battery-powered assets. Empowering battery engineering teams with predictive analytics across their product life cycle will be a crucial competitive advantage to accelerating the scale-up of domestic battery technology development and deployment. Bringing better battery technology to market faster and ensuring a long, safe operating life will, in turn, catalyze the transition away from fossil fuels and towards electric vehicles, grid-scale energy storage, and other clean technologies. The social and economic implications include clean energy jobs, improved environmental quality, and ubiquitous low-cost energy. The potential commercial impact of this project will help accelerate the development and deployment of new battery-powered vehicles, energy storage systems, and other assets. It will allow the company to serve the wider battery industry by de-risking operation and extending service life of battery assets, thereby increasing customer revenue and avoiding costly warranty events. This Small Business Innovation Research (SBIR) Phase II project's goal is to de-risk the deployment, operation, and maintenance of battery energy storage systems. It will combine results from the Phase I with data from partners to forecast system maintenance and inform warranty design, thereby lowering the total cost of ownership and minimizing liability. Access to cell testing, outgoing quality control, and field data will allow for a deep dive across the product life cycle to identify how known degradation mechanisms manifest in the real-world battery data. Physics-informed feature engineering will be used to extend models to incorporate these insights and then implement these models at scale in the cloud. Criteria for success include: 1) correlating real-world operating conditions with known Lithium-ion battery degradation pathways, 2) engineering new features that are correlated with physics- and electrochemical-based insights, 3) accurately estimating remaining useful life to within 5% of total cycle life, and 4) implementing data-driven model in a scalable cloud environment. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.