SBIR Phase I: Cloud-Based Data-Driven Predictive Analytics for Battery Performance

Voltaiq, Inc., Berkeley CA

Investigators

Abstract

This Small Business Innovation Research (SBIR) Phase I project represents a major advance in battery research, combining the use of large, comprehensive battery datasets with advanced data science techniques and cloud-based software architectures to bring an unprecedented level of analytical capability to the problem of modeling battery performance. Past efforts at developing
battery performance models have relied upon manufacturers? performance data and limited single-discharge studies. This innovation applies contemporary data science techniques to the analysis of a
large, normalized set of battery data including raw time-series data and aggregated per-cycle
performance from many battery cycling tests, as well as "lab notebook" data including cell
composition, dimensions, test methods, and observations. The result will be the first commercially-available tool for conducting comprehensive, multi-parameter empirical studies of battery performance.
This unique, data-driven analytical capability will suggest new and promising research paths
distilled from relationships hidden in the data, and will help to predict battery performance
and lifetime. The broader impact/commercial potential of this project is manifested in its potential to dramatically increase the pace of product innovation
and improvement in the battery sector. Organizations developing new batteries and those integrating batteries into their products will perform more targeted and effective battery tests, and
will gain deeper insights from the data more quickly. Furthermore, our predictive analytics
module will become more effective as the total volume of battery performance data stored in
the system increases, further accelerating the pace of development. As this innovation is
applicable to the entire spectrum of battery chemistries and designs, its successful implementation and commercialization will result in improvements in performance and reliability of batteries
and battery-powered devices across a wide range of applications; from smartphones and tablets, to medical devices, to electric vehicles, and grid-scale energy storage. Broad adoption of
this software will accelerate the development and deployment of energy storage and alternative energy technologies, promoting economic growth, energy independence, and environmental benefits. Market research suggests a total addressable market of up to $600M per year for this battery
data platform with advance predictive analytics.