ABB-Siemens - Interoperability (PRP & HSR) Demo at Cigré 2010-MuellerTh

School of
Engineering
DATEN-KARUSSELL MIT
GEGENVERKEHR
Prof. Thomas Müller
Institute of Embedded Systems (InES)
© Slides by HSR project team Interoperability demo at CIGRE 2010

Zürcher Hochschule für Angewandte Wissenschaften School of
Engineering
School of Engineering
Gegründet 1874 als "Technikum Winterthur"
> 1‘300 Studierende in Bachelor Studiengängen
ca. 300 Diplomierte pro Jahr
60 Studierende im MSE Master of Science in Engineering

School of
Institute of Embedded System (InES)

Engineering
Personal:
45 Mitarbeiter total
25 Assistenten und
wiss. Mitarbeiter
10 Dozenten, (davon 2 Teilzeit)
7 technische Mitarbeiter
1 Sekretärin
2 Praktikant / Lehrling
Finanzen 2009:
F&E: 2.8 Mio. CHF
Lehre: 1.9 Mio. CHF

School of
Engineering
Schwerpunkte am InES
Industrielle Echtzeit-Kommunikation:
Realtime Ethernet: Hans Doran
Zeitsynchronisation und Hochverfügbarkeit: Hans Weibel
System on Chip Design: Hans-Joachim Gelke
Wireless Communication: Marcel Meli
Entwicklungsmethoden: Hugo Fierz

Tanks to the
HSR Project Team
for the slides
Prof. Dr. Hubert Kirrmann
ABB
Oliver Kleineberg
Hirschmann
Clemens Hoga
Siemens
Prof. Thomas Müller

ZHAW
CIGRE 2010, Paris
SIEMENS
School of
Content Engineering
1. Short (!) Introduction

2. Standards (IEC 62439) and requirements (IEC 61850)
3. PRP (Parallel Redundancy)
4. HSR (High-availability seamless redundancy)
5. KTI Project
6. Conclusion

School of
Redundancy requirements
Engineering
IEC TC57 WG10 (POWER SYSTEM IED COMMUNICATION

AND ASSOCIATED DATA MODELS)
studied how long the substation automation application can
tolerate the loss of communication without causing a malfunction
of the protection system (next slide).
In substation automation, redundancy fulfilling the (n-1) criteria
(no single point of failure) is crucial
Substation automation systems require communication networks
that fulfill the (n-1) criteria as well
A most important criteria for redundancy is the recovery time:
How long does it take to restore service after a failure

School of
Recovery delay demands as shown in IEC 61850-5 ED2 Engineering
Communicating Service Application recovery Required

partners tolerated delay Communication
Recovery Time
SCADA to IED, client- IEC 61850-8-1 800 ms 400 ms
server
IED to IED interlocking IEC 61850-8-1 12 ms 4 ms

(with Tmin set to 4 ms)
IED to IED, reverse IEC 61850-8-1 12 ms 4 ms

blocking (with Tmin set to 4 ms)
Protection trip excluding IEC 61850-8-1 8 ms 4 ms

Bus Bar protection
Bus Bar protection IEC 61850-9-2 < 1 ms Bumpless

on station bus
Sampled Values IEC 61850-9-2 Less then two consecutive Bumpless
on process bus samples
To fulfill these requirements, IEC 61850-8-1 and -9-2 uses redundancy solutions
standardized for Industrial Ethernet by IEC 62439-3.
IEC 62439: School of
Engineering
High Availability Automation Networks

Ethernet based high availability networks
Specifies relevant principles
Independent of application protocol
Can be used for any application
Industrial automation
Energy automation
Transportation
Medical

School of
Redundancy solutions Engineering
With small recovery time

RSTP (Rapid Spanning Tree Protocol)
IEEE 802.1D-2004
IEC 62439-1 Annex A (recovery time calculation methods)
MRP (Media Redundancy Protocol)
IEC 62439-2
With “zero” switchover time (seamless)

PRP (Parallel Redundancy Protocol)
IEC 62439-3 Clause 4
HSR (High-availability Seamless Redundancy)
IEC 62439-3 Clause 5

School of
RSTP Rapid Spanning Tree (IEEE 802.1D) Engineering
RSTP is a mature protocol widely used in offices, industry and substations.

RSTP can provide a predictable delay under a second in restricted topologies
(especially rings with a limited number of switches) as specified in IEC 62439-1.
RSTP does not cover edge port failures and it provides only a small availability
increase.
root port
{forwarding}
designated port alternate port
{forwarding} station bus {blocked}
P1 P2 P1 P2 root port P1 P2 edge port

{forwarding}
C C C
MU MU MU MU MU MU
bay bay bay

School of
Media Redundancy Protocol (IEC 62439-2) Engineering
Redundancy for Ring Topology with
guaranteed Switch Over Time

School of
Seamless redundancy principles (example PRP) Engineering
Apply two interfaces

Send same information
twice on different paths
Prevent duplication;
discard the second
message
Loop suppression
DAN Dual Attached Node, Keep redundancy

SAN Single Attached Node invisible to the application
RedBox Redundancy Box
VDAN Virtual Dual Attached Node

School of
PRP Parallel Redundancy Protocol Engineering
(IEC 62439-3 Clause 4)
PRP provides (n-1) redundancy (no single point of failure) and

availability increase is high.
PRP requires the full duplication of the network, but only of those
parts that require seamless redundancy.
PRP allows to attach standard Ethernet devices to one of the

LANs. Therefore, networks with redundant and non-redundant
parts can be built.
PRP can be easily implemented in software by a dedicated driver

that manages two standard Ethernet ports.

School of
PRP principle DANP = Doubly Attached Node using PRP

Engineering
“A” frames DANP

“B” frames DANP
standard frames
switch switch
local area network local area network

A B
switch switch switch switch
DANP DANP DANP DANP DANP SAN SAN
Two Ethernet networks (LANs) of similar topology operate in parallel.

The LANs shall not be connected to each other to ensure fail-independence.
Each node is a doubly attached node with PRP (=DANP) which has an interface for each LAN
and sends a frame simultaneously on both LANs.
A receiver receives in normal operation both frames and discards the duplicate.
In case of failure, a receiver keeps on with the frames it receives from the healthy port.
When traffic is reestablished, the receiver resumes processing frames from both channels.
School of
PRP normal operation Engineering
application application
IED IED
Redundancy Redundancy
Manager device interface Manager
A B AA B
Network A
Network B
Redbox (Redundancy Manager in a Box)
IED

PRP normal operation (with long delay on School of
Engineering
Network B)
IED IED
A B AA B
Network A
Network B
Redbox (Redundancy Manager in a Box)
IED

School of
PRP operation with fault Engineering
IED IED
A B A B
Network A
Network B
Redbox
IED

HSR High-availability seamless redundancy IEC School of
Engineering
62439-3 Clause 5
Provides (n-1) redundancy for all network components (no single

point of failure)
Provides seamless redundancy for station bus and process bus
Allows to build rings without using switches
Cost-effective solution
Requires dedicated hardware in the nodes to keep the forwarding
delay low.

School of
Engineering
HSR principle (Multicast)
source destinations
“A” frames
„C“-frame node „D“-frame node node
“B” frames
standard frames
removal from the ring
„A“-frame „B“-frame
(HSR tagged) (HSR tagged)
B A
node node node node node
destinations
Nodes are arranged as a ring, each node has two identical interfaces, port A and port B.
For each frame to send (“C”-frame), the source node sends two copies over port A and B.
Each node relays a frame it receives from port A to port B and vice-versa, except if it already forwarded it.
The destination nodes consumes the first frame of a pair and discards the duplicate.
In case of interruption of the ring, frames still continue to be received over the intact path.

School of
HSR normal operation Engineering
IED IED
Manager
device interface Manager
Redbox
Manager Manager
IED
IED IED

School of
HSR operation with fault Engineering
IED IED
Manager
device interface Manager
Redbox
Manager Manager
IED
IED IED

School of
HSR: rings of rings: three levels Engineering
workstation printer
GPS
HSR
standard Ethernet
quadbox
maintenance laptop
no RSTP protocol any more (but can be used)

note that level 3 is singly attached (only one quadbox)
PRP and HSR are complementary and School of
Engineering
can be connected

PRP and HSR are complementary and School of
Engineering
can be connected
end end
node PRP nodes node
A B A B
PRP LAN A LAN B
interlink A interlink B
end
node
RedBoxPH B
RedBoxPH A
A B A B A B
HSR
B A B A B A B A
end end end end

node node node node

School of
PRP/HSR in Parallel and Ring Redundancy Engineering
application
SC
Redundancy
Manager
IED
Station Bus
PRP or
HSR in parallel mode IED
IED
IED
IED Process Bus

HSR
IED
IED IED
application
School of
PRP/HSR in Parallel and Ring Redundancy Engineering
application
SC
Redundancy
Manager
IED
Station Bus
PRP or Failure in
HSR in parallel mode IED
IED
IED
parallel string
IED Process Bus

HSR
IED
IED IED
application
School of
KTI Project Engineering
HSR is being evaluated and HSR components (FPGA based) are being
developed as the open source in a research project financed by the Swiss
Federal Government (KTI), with participation of:
Zurich University of Applied Sciences /
Institute of Embedded Systems (InES)
ABB
Siemens
Hirschmann
The IP (HVDL Code and C-Source Code)

of this implementation will be available
at 2Q2010
to fair and reasonable terms
The IP will be maintained and supported

by ZHAW at the Institute of Embedded Systems
School of
InES Communication Module Engineering
Field Programmable Gate Array

(FPGA) based SoC-Design
• Cyclone III, 40k Logic Cells

• Softcore Processor NIOS II running
Linux Operating System
• 4 Ethernet Connections 100 B-TX
• Fast Frame Buffer (256k x 36 Bit)
• 4 M Byte Parallel Flash
• 16 M Byte Serial Flash
• 16 M Byte SDRAM
• 24V Power Supply
• Extendable by Piggyback Board

School of
Interoperability Demo at CIGRE Paris Engineering
August 25th-26th 2010
ABB
Tx-Fx
fault generator
Flexibilis
ZHAW
Tx-Fx
RuggedCom
Hirschmann
Siemens

School of

School of

School of

School of
Summary Engineering
IEC 61850 specifies several redundancy methods
PRP seamless, for high-end substations, combines topologies

redundant and not redundant, easily to implement in software
SW-Drivers available from at least two suppliers (ZHAW, NetModule);
RedBox IP (VHDL and Code) for FPGA-Implementation from ZHAW.
HSR seamless, for voltage levels and process bus, rings and rings of rings
Requires nodes with dedicated hardware, but after this initial investment,
provides a cost-effective redundancy.
Allows ring and parallel topologies, but not mixing redundant and non-
redundant devices on the same network (RedBox is required).
Several companies are developing components with the support of the KTI
project of ZHAW.

School of
KTI-Studien und Projekte Engineering
Voraussetzungen:
innovativer / technischer Fortschritt (hohes Risiko)
mindestens ein Wirtschaftspartner
wirtschaftliche Bedeutung in der Schweiz
Projektplan: Wer, Was, Wann
Nutzungsvertrag
Aufwendungen der Hochschulen werden durch KTI finanziert
Anteil bei Machbarkeitsstudien bis zu 80%
Anteil bei Projekten bis 50%

School of
Engineering
InES – Institute of Embedded Systems

Technikumstrasse 9, Postfach
CH-8401 Winterthur
Tel: +41 (58) 934 75 25

http://ines.zhaw.ch
mail-to: info.ines@zhaw.ch

ABB-Siemens - Interoperability (PRP &amp; HSR) Demo at Cigré 2010-MuellerTh

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ABB-Siemens - Interoperability (PRP &amp; HSR) Demo at Cigré 2010-MuellerTh

Uploaded by

Copyright:

Available Formats

School of

Prof. Thomas Müller

Institute of Embedded Systems (InES)

© Slides by HSR project team Interoperability demo at CIGRE 2010

© Slides by HSR project team Interoperability demo at CIGRE 2010

Institute of Embedded System (InES)

10 Dozenten, (davon 2 Teilzeit)

© Slides by HSR project team Interoperability demo at CIGRE 2010

© Slides by HSR project team Interoperability demo at CIGRE 2010

Prof. Thomas Müller

CIGRE 2010, Paris

1. Short (!) Introduction

© Slides by HSR project team Interoperability demo at CIGRE 2010

 IEC TC57 WG10 (POWER SYSTEM IED COMMUNICATION

© Slides by HSR project team Interoperability demo at CIGRE 2010

Communicating Service Application recovery Required

IED to IED interlocking IEC 61850-8-1 12 ms 4 ms

IED to IED, reverse IEC 61850-8-1 12 ms 4 ms

Protection trip excluding IEC 61850-8-1 8 ms 4 ms

Bus Bar protection IEC 61850-9-2 < 1 ms Bumpless

High Availability Automation Networks

© Slides by HSR project team Interoperability demo at CIGRE 2010

Redundancy solutions Engineering

 With small recovery time

 With “zero” switchover time (seamless)

© Slides by HSR project team Interoperability demo at CIGRE 2010

RSTP Rapid Spanning Tree (IEEE 802.1D) Engineering

 RSTP is a mature protocol widely used in offices, industry and substations.

P1 P2 P1 P2 root port P1 P2 edge port

bay bay bay

Media Redundancy Protocol (IEC 62439-2) Engineering

Redundancy for Ring Topology with

guaranteed Switch Over Time

© Slides by HSR project team Interoperability demo at CIGRE 2010

Seamless redundancy principles (example PRP) Engineering

 Apply two interfaces

DAN Dual Attached Node,  Keep redundancy

© Slides by HSR project team Interoperability demo at CIGRE 2010

(IEC 62439-3 Clause 4)

 PRP provides (n-1) redundancy (no single point of failure) and

 PRP allows to attach standard Ethernet devices to one of the

 PRP can be easily implemented in software by a dedicated driver

© Slides by HSR project team Interoperability demo at CIGRE 2010

PRP principle DANP = Doubly Attached Node using PRP

“A” frames DANP

local area network local area network

DANP DANP DANP DANP DANP SAN SAN

Two Ethernet networks (LANs) of similar topology operate in parallel.

PRP normal operation Engineering

© Slides by HSR project team Interoperability demo at CIGRE 2010

© Slides by HSR project team Interoperability demo at CIGRE 2010

PRP operation with fault Engineering

© Slides by HSR project team Interoperability demo at CIGRE 2010

 Provides (n-1) redundancy for all network components (no single

© Slides by HSR project team Interoperability demo at CIGRE 2010

node node node node node

© Slides by HSR project team Interoperability demo at CIGRE 2010

HSR normal operation Engineering

© Slides by HSR project team Interoperability demo at CIGRE 2010

HSR operation with fault Engineering

© Slides by HSR project team Interoperability demo at CIGRE 2010

HSR: rings of rings: three levels Engineering

 no RSTP protocol any more (but can be used)

© Slides by HSR project team Interoperability demo at CIGRE 2010

ABB-Siemens - Interoperability (PRP & HSR) Demo at Cigré 2010-MuellerTh

ABB-Siemens - Interoperability (PRP & HSR) Demo at Cigré 2010-MuellerTh

IEC TC57 WG10 (POWER SYSTEM IED COMMUNICATION

With small recovery time

With “zero” switchover time (seamless)

RSTP is a mature protocol widely used in offices, industry and substations.

Apply two interfaces

DAN Dual Attached Node, Keep redundancy

PRP provides (n-1) redundancy (no single point of failure) and

PRP allows to attach standard Ethernet devices to one of the

PRP can be easily implemented in software by a dedicated driver

Provides (n-1) redundancy for all network components (no single

no RSTP protocol any more (but can be used)

to fair and reasonable terms