Testbed for Mitigation of Power Fluctuation on Micro-Grid

Presented by Xin Zhao

UC San DiegoApril 1, 2015

10th Carnegie Mellon Conference onThe Electricity Industry

Acknowledgements

• Xin Zhao System Identification and Control Laboratory, UC San Diego

• Raymond de Callafon System Identification and Control Laboratory, UC San Diego

• Maurice van de Ven Eindhoven University of Technology

• William Torre Center for Energy Research, UC San Diego

• Chuck Wells Center for Excellence, OSIsoft

UC San Diego Team

• Greg Smedley One-Cycle Control Inc.

• Tong Chen One-Cycle Control Inc.

OCC Team

2 of 20

The project was sponsored by the California Energy Commission.

Outline

• Introduction

• Testbed for mitigation of power fluctuation– Overview– “Portable” cabinet– Controller

• Future work

• Preliminary tests on the testbed

3 of 20

Introduction

Introduction

Motivation for mitigation of power fluctuation

• Power fluctuation occurs intermittently on micro-grid.

• Conventional generation tends to stabilize and maintain synchronous operation of the system by the inertia in the form of spinning rotational mass.

• As more renewable energy generation is added to the utility grid, it could result in instability and poorly damped oscillations in AC frequency and power on micro-grid.

5 of 20

Introduction

Objectives of building a testbed

• Simulate a power fluctuation– Motor load– Oscillatory circuitry

• Detection of instantaneous fluctuations– Phasor Measurement Unit (PMU)– Instantaneous power sensor

• Verification of data-based dynamic modeling (system identification) techniques

• Damping controller design and implementation– Embedded devices

• Capability of real-time control of an inverter– An inverter with real-time active/reactive power control

6 of 20

Testbed

Testbed – Overview

PVSystem

Grid-TiedInverter

EMIFilter

L1

L2

L3

A

Controller

A

V

AuxiliaryRelay

OverloadProtectionContactor

CircuitBreaker

GRID

Sensors

Switch

A

V

V

R-L-C Load Circuit

System diagram

Inverter with real-time control

Oscillatory circuitry acting as a power fluctuation

Controller with implementation of instantaneous power calculation and damping control

8 of 20

Testbed – Key Components

“Portable” cabinet

Oscillatory R-L-C circuitry

Grid-Tied Inverter Controller Cabin

9 of 20

Testbed – Key Components

Controller

Manufacturer: National InstrumentsModel: NI myRIO-1900

• Processor: Xilinx Z-7010 (Duo Core, 667MHz)• Memory: (ROM) 256MB (DDR3) 512MB• Wireless: IEEE 802.11 b,g,n• Analog Input: 12 bits – 500 kS/s• Analog Output: 12 bits – 345 kS/s

10 of 20

Preliminary Tests

Preliminary Tests

Test of simulating power fluctuation

• For safety consideration, a programmable DC power supply is installed for testing.

• The power fluctuation generated by the oscillatory circuitry is measured and modeled as follows:

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3-40

-20

0

20

40

60

80

100

Time [sec]

3-ph

ase

Rea

l Pow

er

System MeasurementModel Simulation

12 of 20

Preliminary Tests

Four-quadrant grid-tied inverter (GTI)

Manufacturer: One-Cycle Control (OCC)Model: GTI3100A6208/3652IR-PQ

• Max. Power: 36kW• AC Voltage Range: 208V ±10%• Rated DC Voltage: 365VDC• Max. AC/DC Current: 100Arms / 100A• Weight: 65lb • Size: 23in×17.5in×5.25in

13 of 20

Preliminary Tests

Capability test of real-time active/reactive power control

• Dynamic response of the OCC-GTI is tested with a step control input.

• The OCC-GTI is capable to be controlled in real time.

14 of 20

4 4.2 4.4 4.6 4.8 5 5.2 5.4 5.6 5.8 6-20

-10

0

10

20

30

40

Rea

l Pow

er

Time (sec)4 4.2 4.4 4.6 4.8 5 5.2 5.4 5.6 5.8 6

0

1

Ste

p In

put t

o G

TI

Preliminary Tests

Verification of data-based system identification on the GTI output

• Based on the measured data obtained by previous tests, a low-order model built within Prediction Error (PE) framework is capable to capture the dynamics.

15 of 20

4 4.2 4.4 4.6 4.8 5 5.2 5.4 5.6 5.8 6-20

-10

0

10

20

30

40

Time [sec]

3-ph

ase

Rea

l Pow

er

Measured ResultSimulated Result

Preliminary Tests

Verification of data-based system identification on the disturbance

• Dynamic response of the oscillatory circuitry is tested.

• A low-order model built by Step-Based Realization (SBR) method is capable to capture the dynamics well.

16 of 20

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2-40

-20

0

20

40

60

80

100

Time [sec]

3-ph

ase

Rea

l Pow

er

Measured ResultSimulated Result

Preliminary Tests

Damping control algorithm design and implementation

• A preliminary damping control algorithm is designed based on modeling of the system described previously.

• The control algorithm is implemented in the controller.

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3-40

-20

0

20

40

60

80

100

Time [sec]

3-P

hase

Rea

l Pow

er

Open-Loop SimulationClosed-Loop Simulation

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2-80

-60

-40

-20

0

20

40

60

80

100

120

Time [sec]

3-P

hase

Rea

l Pow

er

No ControlWith Feedback Control

Control Algorithm Design Control Algorithm Implementation

17 of 20

Preliminary Tests

Conclusions

• The oscillatory circuitry in the testbed is able to simulate a power fluctuation.

• The grid-tied inverter provided by One-Cycle Control is capable to be controlled in real time.

• The controller is able to process the instantaneous power calculation and real-time control.

• The designed control algorithm is able to dampen the oscillation generated by the oscillatory circuitry.

18 of 20

Future Work

Future Work

Large-scale integration tests

• Integration with Phasor Measurement Unit (PMU)

• Integration with photovoltaic (PV) systems

• Large-scale tests on UCSD micro-grid

20 of 20

Thank you