Asee peer logo

Student Project to Develop a Neural Network-based State of Charge Indicator for Primary Batteries

Download Paper |

Conference

2016 ASEE Annual Conference & Exposition

Location

New Orleans, Louisiana

Publication Date

June 26, 2016

Start Date

June 26, 2016

End Date

August 28, 2016

ISBN

978-0-692-68565-5

ISSN

2153-5965

Conference Session

CAPSTONE (SENIOR) DESIGN AND UNDERGRADUATE PROJECTS

Tagged Division

Energy Conversion and Conservation

Page Count

11

DOI

10.18260/p.25923

Permanent URL

https://peer.asee.org/25923

Download Count

184

Request a correction

Paper Authors

biography

Herbert L. Hess University of Idaho, Moscow

visit author page

Herb Hess is Professor of Electrical Engineering at the University of Idaho, where he teaches subjects in He received the PhD Degree from the University of Wisconsin-Madison in 1993. His research and teaching interests are in power electronics, electric machines and drives, electrical power systems, and analog/mixed signal electronics. He has taught senior capstone design since 1985 at several universities.

visit author page

biography

Edward James William Jr Solved Engineering LLC

visit author page

Dr. Edward J William Jr, Ph.D. P.E. is a Doctoral and M.S. graduate of Electrical Engineering from the University of Idaho. In addition, he holds a B.S. graduate of Electrical Engineering at Arizona State University and is a registered Professional Engineer (P.E.). Edward is currently employed by Commonwealth Edison (ComEd) and serves as a Power Systems Protection and Controls (P&C) Engineer. He has offered his P&C Engineering skillsets to Primera Engineers Ltd. an Engineering Consulting firm based in Chicago IL, IPS-Energy based in Munich Germany, and SPX Transformer (Formerly Waukesha Electric Systems) since graduation and has over 5 years of technical engineering experience with The Boeing Company and Honeywell International. His work experience includes design and operation of transformers, power system protection, and substation asset management. He has a special interest in substation protection and control applications. finally, Edward has publications for scientific contributions in IEEE and Cigre Canada.

Edward is the business owner of Solved Engineering LLC, is one of few African American owned Power System Engineering Firm in Chicago. Edward has held National Leadership positions in IEEE Power and Energy Society Standards committees, Eta Kappa Nu (President), and the National Society of Black Engineers (NSBE) (National Executive Officer and Chapter President). Edward is also involved in the IEEE Power and Energy Society, Cook County Forest Preserve Foundation, Chicago’s Young and Powerful, Alpha Phi Alpha Fraternity Incorporated, Rotary International (Rotary One), and is a fellow traveler (Oriental 33).

visit author page

Download Paper |

Abstract

This paper describes how a team of five students, all but one of them undergraduates, successfully developed a State of Charge Indicator (SOCE) for the Li/CFx battery. The pedagogical methods that led to a successful outcome were originally proposed in an ASEE conference paper by the supervising professor nearly a decade ago.

Improving the accuracy of a SOCI for the Li/CFx battery is difficult due to the flat discharge profile of the battery and its non-linear response to ambient temperature. To account for these effects, an Artificial Neural Network (ANN) was designed to run on the MSP430 microcontroller. The ANN was developed and trained with data acquired from a mathematical model and laboratory testing of a Li/CFx cell. The ANN uses voltage, current, and ambient temperature for its inputs, and outputs the State of Charge (SOC) of the cell in the form of a five LED display. For military use, difficult constraints on temperature, power consumption, cost, and size were imposed.

Understanding the battery is the first task. Students performed discharge curves on a number of Li/CFx cells. From these tests, the flat discharge profile challenge became evident. A charge counter algorithm was selected, modified for temperature, rate of discharge, amount of discharge, power consumption, and battery history. Finding a simple charge counter in the technical literature, a circuit which nearly fit under a dime, a first version SOCI prototype was developed which successfully monitored voltage, current, and ambient temperature. Subsequent prototypes improved on accuracy, power consumption, and cost.

A team of five students, all but one of them undergraduates, worked on this project and learned from it for over 30 months. Assessment of their learning will follow up on the methods proposed by one of the authors nearly a decade ago. Assessment also includes quality of positions found by the students: at SpaceX, Agilent, Orbital Technologies, to an engineering faculty position, and to a large electrical equipment manufacturer. The success led to a reputation for quality work that secured follow-on Defense projects as will be described in the paper.

Hess, H. L., & William, E. J. (2016, June), Student Project to Develop a Neural Network-based State of Charge Indicator for Primary Batteries Paper presented at 2016 ASEE Annual Conference & Exposition, New Orleans, Louisiana. 10.18260/p.25923

ASEE holds the copyright on this document. It may be read by the public free of charge. Authors may archive their work on personal websites or in institutional repositories with the following citation: © 2016 American Society for Engineering Education. Other scholars may excerpt or quote from these materials with the same citation. When excerpting or quoting from Conference Proceedings, authors should, in addition to noting the ASEE copyright, list all the original authors and their institutions and name the host city of the conference. - Last updated April 1, 2015