Paul Cronk

University of Minnesota

October 14th, 2015

Hydraulic Flywheel Accumulator for Mobile Energy Storage

2

OutlineI. Overview

I. Background on Mobile Energy Storage

II. Hydraulic Flywheel Accumulator Concept

III. Research Goals and Progress

II. Modeling

I. Architecture

II. Hydraulic Losses

III. Kinetic Losses

III. Prototype and Experimental Setup

IV. Preliminary Experimental Results

V. Future Work

3

OVERVIEW

4

Why Hybridize a Powertrain?

• Recapture kinetic energy otherwise dissipated as heat

during braking events

• Downsize engine and run at peak efficiency point without

compromising vehicle performance

• With “plug-in” capability, utilize cleaner energy sources

for vehicle propulsion

Axle

Pump/Motor

Pump

Engine

Accumulator

5

Hydraulic vs. Electric Powertrain

Hydraulic Powertrain Electric Powertrain

Component Weight

Component Cost

ESS Lifetime

ESS Power Density

ESS Energy Storage Density

Li-Ion Battery Energy Density: 432𝑘𝐽

𝑘𝑔Hydraulic Accumulator Energy Density: 6

𝑘𝐽

𝑘𝑔

6

0 0.5 1 1.5 2 2.5 3 3.5 40

5

10

15

20

25

30

35

40

45

50Dimensionless Pressure vs. Dimensionless Energy

Energy [E/(Pcharge

Vcharge

)]

Pre

ssure

[P

/Pcharg

e]

Pressure vs. SOC

In a traditional accumulator, pressure varies with state-of-charge.

7

Flywheel-Accumulator Concept

• Rotating Pressure Vessel

• Piston Separates Compressed

Gas and Oil

• Torque is applied at the gas side

• Two Energy Storage Domains:

– Hydro-Pneumatic

– Rotating Kinetic

8

Analyzing Fluid Pressure

FBD of Infinitesimal Volume:

sPA

dθ

2 odPP A

2 idPP A

sPA l

dr

X

Y

r

2 2

2S

rP r P

Influences on System Pressure:

• Adding Oil Increases Pressure

• Increasing Angular Velocity Decreases Pressure

9

Flywheel-Accumulator Concept

• Pressure rises as flywheel kinetic energy is extracted

• More pneumatic energy is stored for the same system pressure.

• Less kinetic energy is required to maintain pressure.

• Potential for higher efficiency than a spatially separated combination of

flywheel and accumulator

2 2

2S

rP r P

10

Research Goals

• Specify physically feasible design

• Model performance and optimize design

parameters to maximize energy capacity

and efficiency

• Build and test medium energy prototype

• Refine model and for higher energy levels.

11

MODELING

12

Steel liner with circumferential composite filament winding

– Composite provides high hoop strength

– Liner ensures piston sealing and alleviates radial tensile stresses in

composite

Architecture

13

– Axial and radial ports in the axle facilitate transport of oil

– The axle, housing, and end caps are radially unconstrained to one another

– Retainers and radial pins provide axial and tangential constraints between

components

Architecture

14

Hydraulic Losses

• Axle Port Losses

• HSRU Leakage

15

Kinetic Losses

• Bearing losses

• Aerodynamic drag

• HSRU Viscous Dissipation

• Pump/Motor losses

– Use modified McCandlish and Dorey model

– Assume one set of loss coefficients provides roughly accurate loss

estimates for a range of PM sizes

• Spin-Up Losses• Strohmaier, K.G. et al., 2014. Experimental Studies of Viscous Loss in a Hydraulic

Flywheel Accumulator. In 2014 Proceedings of the 52nd National Conference on

Fluid Power.

16

Optimization Method

Energy density

Drive cycle efficiency

0 200 400 600 800 1000 1200 14000

0.2

0.4

0.6

0.8

1

time (s)

Vehicle Velocity and State-of-Charge vs. Time

state-of-charge

vehicle velocity (normalized)• Multi-Objective Genetic

Optimization

17

2 4 6 8 10 12 14 1680

82

84

86

88

90

92

94

96

Ed (

W-h

)

ud (kJ/kg)

2 4 6 8 10 12 14 1620

30

40

50

60

70

80

90

100

msys (

kg)

Ed

msys

Selection of a Prototype Design

• Heavy solutions incur

high rolling resistance

• Very energy-dense

solutions incur high

losses

• Minimize Wdc

minimize Ed

System mass, 𝒎𝒔𝒚𝒔 39.3 kg

Energy density, 𝒖𝒅 8.77 kJ/kg

Energy capacity, 𝑬𝒅 81.8 W-h

Mass, excluding PMs 32.3 kg

Capacity ratio, 𝑹𝒄 76.3

Housing safety factor 7.35

Storage PM displacement, 𝑫 0.63 cc/rev

Packaging volume (approx.) 48.6 liters

Drive cycle losses, 𝑾𝒍𝒐𝒔𝒔 79.2 kJ

Drive cycle efficiency, 𝜼 86.8 %

Usage ratio, 𝑹𝒖 1.97

Pressure fraction, 𝒇𝒑𝒓𝒆𝒔𝒔𝒖𝒓𝒆 48.0 % (+26.8 % / -21.2 %)2 4 6 8 10 12 14 16

80

82

84

86

88

90

92

94

ud (kJ/kg)

(

%)

18

PROTOTYPE

AND

EXPERIMENTAL SETUP

19

Prototype Components

20

Chamber and Drive Section

21

Test Circuit

22

PRELIMINARY

EXPERIMENTAL RESULTS

23

Flywheel Mechanical Efficiency

𝜂𝑚,𝑝 =𝑇𝑖𝑑𝑒𝑎𝑙𝑇𝑓𝑤

=𝑇𝑖𝑑𝑒𝑎𝑙

𝐼𝑑𝑒𝑠𝑖𝑔𝑛𝛼𝑓𝑙𝑦𝑤ℎ𝑒𝑒𝑙𝜂𝑚,𝑚 =

𝑇𝑓𝑙𝑦𝑤ℎ𝑒𝑒𝑙

𝑇𝑖𝑑𝑒𝑎𝑙=𝐼𝑑𝑒𝑠𝑖𝑔𝑛𝛼𝑓𝑙𝑦𝑤ℎ𝑒𝑒𝑙

𝑇𝑖𝑑𝑒𝑎𝑙

24

Hydraulic Pump/Motor Volumetric

Efficiency

𝜂𝑣 = 𝑉𝑎 𝑉𝑖 𝜂𝑣 = 𝑉𝑖 𝑉𝑎

26

Rotary Union Leakage

𝑊𝑙 =𝑃𝑠2𝜋𝑑𝑠𝑐𝑠

3

12𝜇𝑜𝑙𝑠

27

Future Work

• Explore loss mechanisms at higher flywheel speeds

• Explore the effect of fluid spin-up on HFA performance

• Implement the HFA prototype in a simulated drive cycle

• Use validated models to explore benefits of scale

28

Thank You