Battery Life Estimation

Rising gasoline and diesel fuel prices have resulted in a resurgence of interest in hybrid electric and plug-in hybrid electric vehicles. The U.S. Advanced Battery Consortium and the U.S. Department of Energy (DOE) established a goal of 15 years and 150,000 miles for battery development for these applications. However, no information is currently available demonstrating these goals. The key question is how battery performance will change with calendar age, cycles, internal component aging, cell-to-cell manufacturing variations, and summer and winter temperature extremes, among other factors. In addition, a number of new battery features, including advanced coatings for anodes and cathodes, and electrolyte variations and additives, are expected to be included in new battery designs. From the practical point of view, these questions need to be answered by using only limited data.

The Battery Life Estimation (BLE) Software, is a state-of-the-art tool kit for fitting battery aging data and for battery life estimation. It was designed to make life-cycle estimates using, at most, two years worth of aging data.

Applications and Features

The BLE software can assist in answering engineering questions. The software employs a generalized statistical approach to fit data from accelerated aging experiments to a life equation. No specific life equation is assumed; you must enter your own technology-specific equation. If a specific equation is not available, the authors have included a life equation that assumes parabolic and Arrhenius-like kinetics. Included is a linear error model to estimate cell-to-cell variability and measurement errors. A two-sigma error bound is calculated for the variance. The BLE software is different from other curve-fitting routines as it employs robust fitting techniques and has the ability to estimate battery life by using Monte Carlo techniques (which most generalized curve-fitting software does not consider). A Monte Carlo simulation is available to assess lack-of-fit and uncertainty limits for average cell life. The Monte Carlo simulation is only used for the 15-year predictions. The software can operate in three modes:

1. Data fit only,
2. Data fit and Monte Carlo simulation, and
3. Monte Carlo simulation only.

The BLE software has an easy-to-use graphical user interface and a user's manual (expected to be completed in December 2009) with examples and a frequently asked questions section. The software is relatively easy to use and would take 20 to 60 minutes for a user to become familiar with the code. Run times for the data fitting routines are very fast. Monte Carlo predictions take a little longer but are generally complete in about 2 minutes. It operates on a Windows 2000 or later system and uses NET framework extensions, which are widely available at no cost from Microsoft. The software can run on a standard PC.

Licensing Information

To license the BLE. software from Argonne, print a copy of the End-User Software Agreement, arrange for signature by your organization, and send the signed license agreement to:

Aaron Sauers
Software Licensing Coordinator
Technology Development and Commercialization
Building 201
Argonne National Laboratory
9700 S. Cass Ave.
Argonne, IL 60439

Please contact Aaron Sauers by phone at 630-252-7878 or e-mail to alert him that you are sending the agreement, signed by your organization.

Payment can be made by bank wire, purchase order, or credit card, and an invoice will be provided once the number of copies and payment type has been identified.