US-wide ExoGENI network (PoP at RENCI) RTDS-WAMS testbed at NC State

Controlling Next-Generation Electric Power Grids using High-Speed Cloud ComputingJianhua Zhang, Matthew Weiss, Aranya Chakrabortty and Yufeng Xin

Introduction

Architecture of Wide-Area Control with Cloud-in-the-Loop ExoGENI-WAMS Testbed

Power System Models and Wide-Area Control

• Phasor Measurement Units (PMU) of any utility firstsend data to local phasor data concentrators (PDCs)through a virtual private network (VPN).

• The PDCs then route the PMU data-streams to virtualmachines (VM) in a cloud computing platform such asExoGENI.

• The VMs belonging to each company may form a sub-cloud of their own.

• Each VM then communicates to other VMs, both inside(intra-area) and across (inter-area) the sub-clouds, over amulti-hop Software Defined Network (SDN) such asInternet2 to compute the wide-area control signals in adistributed fashion.

`

Area 1

Area 2 Area 3

Area 4

Power Transmission

NetworkLocal PDC 1

)(3 tx

)(2 tx

)(1 tx

)(4 tx

)(5 tx

)(6 tx)(7 tx

)(9 tx

Local PDC 2

Local PDC 3

)(8 tx

)(10 tx

Internet of Clouds

VM

VM

VM

Local cloud in Area 1

VM

VM

VM

Local cloud in Area 4

Local cloud in Area 2

)(1 tx

)(2 tx

)(3 tx

VM VM

)(4 tx )(5 tx

Local cloud in Area 3

VM VM

)(6 tx )(7 tx

Local PDC 4

)(8 tx

)(9 tx)(10 tx

7 ( )u t4 ( )u t

3( )u t

2 ( )u t

1( )u t

6( )u t

10( )u t

9 ( )u t

8( )u t

5( )u t

Control Signals

Control Signals

Measurem

ent Signals Control Signals

Control Signals

Measurem

ent Signals

With increasing trends in renewable penetration, smart loads, electric vehicles, etc. the USgrid is becoming more vulnerable to instabilities. The Wide-area Measurement Systems(WAMS) technology using GPS-synchronized Synchrophasor measurements is an ideal wayto control such instabilities and blackouts. The challenge, however, is the bottleneck in datacommunication and computation, which cannot be handled by today’s Internet. In this project,we propose the use of cloud-computing networks such as GENI to combat this challenge.

Case Study on US West Coast Grid Model

Architecture

ExoGENI Cloud Computing Platform and Integration

RTDS-PMU based WAMS

GPSRTDS

SEL-421

SEL-

487E

GTNETx2

Switch

ExoGENI

Swing + Excitation Dynamic Equations�̇�𝛿 = 𝜔𝜔𝑖𝑖𝑀𝑀𝑖𝑖�̇�𝜔𝑖𝑖 = 𝑃𝑃𝑚𝑚𝑖𝑖 − 𝐸𝐸𝑖𝑖2𝐺𝐺𝐿𝐿𝑖𝑖 − 𝐷𝐷𝑖𝑖𝜔𝜔𝑖𝑖 − �

𝑘𝑘=1,𝑘𝑘≠𝑖𝑖

𝑛𝑛

𝐸𝐸𝑖𝑖𝐸𝐸𝑘𝑘(𝐺𝐺𝑖𝑖𝑘𝑘 cos 𝛿𝛿𝑖𝑖 − 𝛿𝛿𝑘𝑘 − 𝐵𝐵𝑖𝑖𝑘𝑘 sin(𝛿𝛿𝑖𝑖 − 𝛿𝛿𝑘𝑘))

𝜏𝜏𝑖𝑖�̇�𝐸𝑖𝑖 = −𝑥𝑥𝑑𝑑𝑖𝑖𝑥𝑥𝑑𝑑𝑖𝑖′ 𝐸𝐸𝑖𝑖 + 𝑈𝑈𝑖𝑖 +

𝑥𝑥𝑑𝑑𝑖𝑖−𝑥𝑥𝑑𝑑𝑖𝑖′

𝑥𝑥𝑑𝑑𝑖𝑖′ (�

𝑘𝑘=1

𝑛𝑛

𝐾𝐾𝑖𝑖,𝑘𝑘𝐸𝐸𝑘𝑘 cos 𝛿𝛿𝑖𝑖 − 𝛿𝛿𝑘𝑘 )

GENI Experimenter Tools (omni) Other GENIResources

Other GENIResources

ExoGeni Experimenter Tools (Flukes, NDLLib)

PMU-based WAMS at NC State

VLAN904

SoftwareLayer

PhysicalLayer

Design of damping controller

𝑥𝑥𝑇𝑇𝑄𝑄𝑥𝑥 = |∆𝛿𝛿(𝑡𝑡) |∆𝜔𝜔(𝑡𝑡) ∆𝐸𝐸(𝑡𝑡)𝑄𝑄𝛿𝛿 0 00 𝑄𝑄𝜔𝜔 00 0 𝑄𝑄𝐸𝐸

∆𝛿𝛿(𝑡𝑡)∆𝜔𝜔(𝑡𝑡)∆𝐸𝐸(𝑡𝑡)

𝑢𝑢𝑖𝑖 = �𝑗𝑗=1

5

𝐾𝐾𝑖𝑖,𝑗𝑗𝛿𝛿𝑗𝑗 + �𝑗𝑗=1

5

𝐾𝐾𝑖𝑖,𝑗𝑗+5𝜔𝜔𝑗𝑗 + �𝑗𝑗=1

5

𝐾𝐾𝑖𝑖,𝑗𝑗+10𝐸𝐸𝑗𝑗

Note: K is a full matrix, this indicates that every VM will needto communicate with every other VM to exchange stateinformation.

V3∠θ3

V4∠θ4

jx12

V5∠θ5

V1∠θ1

r12

jx3 P3

E2∠δ2

E3∠δ3

E1∠δ1

E4∠δ4

E5∠δ5

ASG3

Montana,Wyoming.Utah

Canada,Washington,

Oregon

Arizona, NevadaNew Mexico

San Francisco

Los Angeles, BajaCalifornia, Mexico

WECC Transient Response, PSS vs WAC

Case I – Local vs Global Control

WECC Transient Response, RSCAD vs ExoGENI

Case II – Simulated Network vs ExoGENI

Fig.1- Envisioned architecture of wide-area communication

Funded by NSF projects CNS 1531099, 1531047, and 1531061