CAEBAT Battery Thermal Management Project by General Motors, ANSYS and ESim

Presenter: Sandeep Sovani, Ph.D., ANSYS USA

Based On Work of: Taeyoung Han, Ph.D., General Motors

Lewis Collins and Dimitrios Tselepidakis, Ph.D., ANSYS Inc Prof. Ralph White, ESim and University of South Carolina Gi-Heon Kim, Ph.D., National Renewable Energy Lab, USA

Battery is the Key to Vehicle Electrification • 2012 DoE EV Everywhere Grand Challenge goal: make plug-in

electric vehicles as affordable and convenient as today’s gasoline-powered vehicles by 2022

Image credit: “FY 2013 Annual Progress Report”, Energy Storage R&D, Vehicle Technologies Office, U.S. Department of Energy

Engineering Challenges in Battery Development

• Cost • Performance (power and energy density) • Durability and service life (in disparate

environments and drive cycles) • Safety (tolerance to abusive conditions)

• Complex multi-scale, multi-physics system • Rapidly evolving materials and design concepts • Existing software tools not “tuned” for batteries

Thermal

Elec-trical

Chemical

Fluid

Molecular Particle Electrode Cell Pack Vehicle

Central Importance of Thermal Management • Performance and life sensitive to cumulative thermal history • Minimize intra-cell and inter-cell

temperature variations

10

100

1000

ICE Battery

°C (log scale)

Comparison of Thermal Headroom

40 Worst-case ambient Accelerated degradation

Thermal runaway

Alloy melting

Exhaust gas

CAEBAT Project

• Sponsor: U.S. Dept of Energy (DoE) Vehicle Technologies Office – High-risk/high-reward R&D to advance transportation

technologies • Computer Aided Engineering of Electric Drive Vehicle BATteries

– Improved simulation is vital to accelerate design innovation

• Goal: develop and validate more accurate, efficient, modular, flexible, accessible, battery-oriented software tools

• Collaborative team project 2011-14

Project Objectives Faster design cycles and optimize batteries (cells and packs) for improved performance, safety, life, and low cost.

Ability to provide trade off studies between various cooling concepts and the battery pack life.

Various cooling

concepts

Liquid cooling

Air coolingCold plate

Battery Power Profile

Vehicle Driving Cycle

Vehicle Simulator

cooling cost vs. warranty cost

•Address multi-scale, multi-physics interactions

•Provide flexibilities

•Expandable framework

•Validate models

6

Project Technical Approach

• Blend field (CFD) simulation, system simulation, novel scale-coupling and reduced-order methods

• Leverage existing building-blocks – ANSYS Workbench, Fluent, Simplorer – Interoperability via Open Architecture Software (OAS)

Cell Model(field simulation)

Reduced Order Model

Pack Model(system simulation)

Project Technical Approach (continued)

• Address gaps in methods and models (NREL, ESim)

• Experimental validation (GM)

• Sustainable deployment to industry (ANSYS)

ANSYS BATTERY DESIGN TOOL (ABDT)

Field Simulation (“Cell Level”)

System Simulation (“Pack Level”)

Reduced-Order Models (ROM)

Workbench Framework and UI

templates templates OAS files

Simplorer UI

Project Framework

ABDT is the “umbrella” over all capabilities, including the graphical user interface (UI) that automates/customizes battery simulation workflow, leveraging ANSYS commercial products. 9

Tool Validation

Temperature Profiles

Validation of Full Field Simulation

TC01

TC02 TC03

Cell 1 Cell 2

Bus bar

Fin Plate

Cooling Channels

24 cell prototype

A 24 cell module validation test set up for full field simulation against test data for high-frequency pulse charge-discharge.

11

Top View

Validation for 24 Cell Module

Measurement Prediction

No

data

No

data

Coolant

in Coolant

out

Maximum difference between the prediction and the measurement is within 1 OC. 12

Full Field Simulation

13

Unit

Module

Pack

Electric Circuit

System Simulation of 24 Cell Module ABDT User Interface

Generate module and pack model

automatically

Thermal Circuit

LTI ROM System-Modeling Approach for Battery Thermal Simulation

14

LTI = Linear Time Invariant ROM = Reduced Order Model

0.3oC

15

Tem

pera

ture

(o C)

Time (sec)

1 sec pulse (3.5 C-rate) at 50% SOC

Prediction Test data for average battery temperature

Validation of System Simulation

CPU runtime for system simulation is less than 1 minute (Dell Z800 PC) Compared to the full field simulation that takes 4 ~ 5 days with 64 processor HPC

Validation of System Simulation for USO6 Drive Cycle

Comparison Between Simulation and Test Data for US06 Drive-Cycle

Summary: Sustainable Software Tool - ADBT

•Modular: —Integrate physics and chemistry in a

computationally efficient manner. •Provide Flexibilities: —Provide a platform to enable various

simulation strategies. •Provide Expandable Framework: —Enable future users to easily add new

physics of interest. —OAS-compatible. •Validated : —Ensure model predictions agree with

experimental data by performing carefully designed experiments.

•Easy to use

Support of DOE CAEBAT The automotive industry requires CAE design tools that include the following capabilities.

Contact: Sandeep.Sovani@ansys.com

Friday, October 03, 2014 2014 Automotive Simulation World Congress 19

Thank you!