1 | Corporate Presentation, September 20, 2016

Hardware-in-the-loop (HIL) Test of Demand as Frequency controlled Reserve (DFR)

QIUWEI WU/Associate ProfessorCenter for Electric Power and Energy (CEE)

Technical University of Denmark (DTU) 15th September 2016

Outline• PowerLabDK

• Wind power in Denmark

• DFR control logics

• Heat pump model

• HIL test

• Conclusions

PowerLab Combines Experimental Facilities in a Unique Platform

Stakeholders:

Supported by:

Investment:18 million Euro

TECHNOLOGY DEVELOPMENT TESTING TRAINING DEMONSTRATION

Lyngby & Ballerup Campus Risø Campus

Flexible multi-purpose laboratories

The Island Bornholm

Large-scale test system

Full-scale real-lifeenergy system

40,000 people33% wind power50% renewable energyIslanding capability

Bornholm - Full-scale Living Laboratory with 40,000 Inhabitants and 50% Renewable Energy Penetration

Bornholm

Resources:- Wind power- Biomass - Biogas- District heating- Combined heat and power- Solar power- eMobility- Active demand

Features:- Nord Pool market- Islanding capability

TECHNOLOGY DEVELOPMENT TESTING TRAINING DEMONSTRATION

Intelligent Control Lab - Test-bed for Integrated Experiments

Grid in Lab

w. 22LabCells

Full-scalecontrol room

Real-time digital simulator

Power amplifier 150 kW (1.2 MW)

SCADAsystem

Models

Power hardwaretest in lab

Onlinedata

LabGrid with 21 LabCells

Onlinecontrol/

data

Online control/data

Power

Power

I/O

Blade Center(power market/controllers)

Online control/data

Real worldControl hardware

test in labI/O

Outline• Intelligent control lab of PowerLabDK

• Wind power in Denmark

• DFR control logics

• Heat pump model

• HIL test

• Conclusions

Year 2014Danish wind power generation: 39.1% of the electricity consumption

January 2014Danish wind power generation: 63.3% of the electricity consumption

December 21th 2013Danish wind power generation: 102% of the electricity consumption

Single hour July 9th 2015Danish wind power generation: 140% of the electricity consumption

March 11th 2014only 9 MW wind power generated out of installed 4,900 MW

but 480 MW out of 580 MW solar units supplied the grid

Wind Power in Denmark

0

1000

2000

3000

4000

5000

6000

7000

MW

h/h

201225% wind power

DemandWind Power

202050% wind power

DemandWind Power

Wind Power in Denmark

Outline• Intelligent control lab of PowerLabDK

• Wind power in Denmark

• DFR control logics

• Heat pump model

• HIL test

• Conclusions

DFR Control Logic

The DFR control logic type I disconnects and reconnects electric appliances to the grid when the system frequency falls below and recovers above , respectively

offf

onf

DFR Control LogicThe DFR control logic type II is customized for switching thethermostatically controlled loads by adjusting the nominaltemperature set points and .normal

lowT

0( )normalhigh highT T kf f f= + −

0( )normallow lowT T kf f f= + −

normalhighT

Outline• Intelligent control lab of PowerLabDK

• Wind power in Denmark

• DFR control logics

• Heat pump model

• HIL test

• Conclusions

Heat Pump Model• The dynamics of a direct air heating system can be sufficiently

described by three thermal masses.• The ambient air of the building interior has smaller storage

volume, and defines a faster dynamics of the system.• the larger storage volume of the building envelope, or structure,

describes a slower dynamics of the system

( ) ( )1 1i e i a i H w s

i ie ia

IT T T T T Q AC R R

= − + − + + Φ

( ) ( )1 1e a e e e s

e ea iei

IT T T T T AC R R

= − + − + Φ

Outline• Intelligent control lab of PowerLabDK

• Wind power in Denmark

• DFR control logics

• Heat pump model

• HIL test

• Conclusions

HIL Test

HIL Test

00:00:00 00:10:00 00:20:00 00:30:0019.5

20

20.5

21

o C

On/off control

Tinside

Tstructure

00:00:00 00:10:00 00:20:00 00:30:000

5

10

kW

Pelectricity

Pheat

00:00:00 00:10:00 00:20:00 00:30:004

5

6

Time (HH:MM:SS)

o C

Toutside

HIL Test

HIL Test

(a)

(b)

Outline• Intelligent control lab of PowerLabDK

• Wind power in Denmark

• DFR control logics

• Heat pump model

• HIL test

• Conclusions

• The DFR technology has been developed to utilize the demandside resources to provide fast reserves needed in the futurerenewable based power system.

• The DFR technology has been tested by offline simulations in theprevious work.

• The real time HIL tests were conducted to verify the effectivenessof the DFR technology.

• The HIL test results show that the DFR technology cansuccessfully arrest the system frequency and illustrate the efficacyof the SmartBox.

Conclusions

21 | Corporate Presentation, September 20, 2016

Thank you for your attention