(Nam-Ho Lee)
1
(Nam-Dae Kim)
1
(Yu-Yeong Park)
†
-
(Senior researcher, Next Generation Transmission & Substation Lab, KEPCO Research Institute,
Korea)
Copyright © The Korean Institute of Illuminating and Electrical Engineers(KIIEE)
Key words
Digital Substation, GOOSE, IEC 61850, IED, Substation Automation System, Verification
1. Introduction
IEC (International Electrotechnical Commission) has released the international standard,
IEC 61850 for data model and communication interfaces which results in implementation
of the substation automation system all over the world with network and Software technologies.
This substation automation system in the center of environmental change of the modern
substation is able to provide and share abundant information under the network system,
such as data for maintenance and diagnosis of power facilities, and for monitoring,
protection and control related to system operation. According to the reason, the conventional
analog type substation is replaced to digital substation using computer based operation
system and IEDs (Intelligent Electronic Device) based on IEC 61850.
This new standard describes MMS(Manufacture Message Specification) and GOOSE(Generic
Object Oriented Substation Event) as the communication services for a server and a
client of the digital substation, and most of utilities that are running the substation
automation system, applied these communication function to operate their digital substation.
They want to verify the new digital interfaces of the IED and operating system and
ask vendors to summit IEC 61850 conformance certificate on both operating system (IEC
61850 client) and IED (IEC 61850 sever) as minimum requirement(1-4).
UCA IUG (International Users Group) published the test procedures of IEC 61850 conformance
test to verify communicative functions and interoperability of the device, but it
does not assure completed system operation at real site because this test is only
the functional test(8).
KEPCO, the public utility in South Korea, has built in about 22 digital substations
since 2013 and plan to construct 100 substations by 2020. And KEPCO Research institute
has been performing technical support as a home doctor whenever any trouble occurs
in the digital substations. During the technical support, A GOOSE trouble was found
at “GOZAN” S/S one of the KEPCO digital substations, which we did not see at other
substations with the same system. So we tried to solve this trouble to run our system
more securely as soon as possible. This paper presents the specific way to find the
reason of the GOOSE trouble found in the substation and introduces how to solve the
issue.
2. GOOSE Monitoring way of KEPCO substation
The communication way for signals in the substation shifts from analog step to digital
step. At once, the legacy substation is using hard-wired cables for electrical signal
based communication between relays and RTU. Otherwise, digital substation is using
specific communication message like GOOSE and MMS. IED and HMI (or Gateway) can communicate
each other based on MMS in order to operate substation automation system with client
and server relation. GOOSE might be used to apply for IED communication. GOOSE is
much faster communication message than MMS owing to three layers(L1, L2 and L7) shown
in Fig. 1(5-6). L2 based network system in the station bus is using Mac tables to switch communication
packets within the network of the digital substation. But if the network switch receives
a packet that is not managed in the table, it must broadcast the packet to all network
switches. Because an IED transmits GOOSE messages to the network only with virtual
Mac address for using the broadcast feature of layer 2 based network switches, transfer
time of GOOSE can be reduced.
Fig. 1. Communication mapping of IEC 61850
Unlike MMS with TCP/IP layers, the GOOSE communication could not be aware of subscribing
this message on other devices. Accordingly, GOOSE should be retransmitted with interval
times in order to compensate the possibility of packet loss. IEC 61850 8-1 shows transmission
time for events like Fig. 2 and explains the meaning of each time as follows(7).
Fig. 2. Transmission time for Events: IEC 61850-8-1
-T0 retransmission in stable conditions (no event for a long time).
(T0) retransmission in stable conditions may be shortened by an event.
-T1 shortest retransmission time after the event.
-T2, T3 retransmission times until achieving the stable conditions time
Fig. 3 shows an example of the GOOSE message captured in the digital substation. As mentioned
above. There is no relationship among GOOSE messages to continue to next communication.
TAL (Time Allowed to Live) shown in Fig. 3 might be used for an IED that is subscribing GOOSE from another IED to expect next
GOOSE message. In other words, the IED should send the next GOOSE message within TAL.
KEPCO has applied GOOSE communication to interlocking signals for switch control of
the digital substation and has implemented substation automation system with IEDs
supporting IEC 61850 Edition 1.0. This means that there is no standardized way to
monitor GOOSE subscription like LGOS of IEC 61850 Edition 2.0. We made a simple logic
of GOOSE subscribed IED that can check whether the IED receive GOOSE messages continuously
or not, as shown in Fig. 4. HMI or Gateway like Fig. 5 can keep an eye on GOOSE communication by receiving MMS report with GGIO including
GOOSE subscription information of the IED, and give us an alarm message if the IED
fails to subscribe GOOSE messages.
Fig. 3. An example of a GOOSE message
Fig. 4. An example of KEPCO GOOSE application
Fig. 5. GOOSE monitoring application of KEPCO digital substation
3. GOOSE trouble Occurrence and Verification
3.1 Occurrence of GOOSE subscription failure
KEPCO currently operating about 22 digital substations has got used to configure the
digital substation, due to the nature of public institutions, involved in a number
of manufacturers.
At “GOZAN” S/S One of the KEPCO digital substations, a transmission protective IED
(hereafter T/L IED) has caused a GOOSE communication trouble due to unknown reasons.
A busbar protective IED (hereafter BUS IED) and others receiving this GOOSE inform
HMI and Gateway the phenomenon failed to subscribe the GOOSE via MMS report as mentioned
above.
Fig. 6. T/L Protection IED in “GOZAN” substation, manufacture: Sanion (South Korea)
The T/L IED is to use the method that is gradually increasing the GOOSE transmission
intervals of 2-4-8 ms in case of an event occurrence. Most of the IEDs that receive
the GOOSE normally are waiting to receive the next within twice time of the TAL. But,
in this case, IEDs such as BUS IED received second GOOSE message after much delay
later than 2ms TAL of the first GOOSE message that T/L IED generated. They announce
the failure reports via MMS.
3.2 How to verify GOOSE transmission performance
In order to find the cause of the GOOSE subscription failure, we analyzed the problem
with the communication function of the IED with the maker but found no special reason.
In addition, the same product is installed on other substations but the phenomenon
such as “GOZAN” S/S does not occur. So the cause of GOOSE failure was decided to look
at the network.
With the assumption that generating GOOSE of T/L IED transmission was normal behavior,
we have configured the field verification by implementing a verification system as
shown below (Fig. 7) in order to determine whether the delay of the GOOSE transmission in the SA network
of GOZAN substations occurs.
Fig. 7. Verification system for GOOSE subscription failure analysis
The verification system had the GOOSE simulator (we called GOOSE Agent) recognized
T/L IED's CID file to simulate GOOSE the same as T/L IED. KEPCO has developed GOOSE
agent to verify GOOSE function on IEDs in the digital substation and Fig. 8 shows GUI of the tool.
Fig. 8. GOOSE agent as GOOSE monitor and simulator
In order to avoid collision of GOOSE communication between the simulator and the IED,
destination Mac and GoID of GOOSE that the GOOSE simulator generates, were changed.
A total of three network analyzers (WireShark application) were used to measure the
transmission delay interval. First it is installed on the same laptop of the GOOSE
simulator (GOOSE network transmission reference point), and the second analyzer is
installed on the network switch connected to the GOOSE simulator through the network
tapping equipment for mirroring communication traffic (the transmission time measurement
of a switch inside connected to T/L IED), the third analyzer is installed on the PC
where HMI server is being operated (the transmission time measurement between network
switches of GOZAN substation).
3.3 Solution on the GOOSE subscription failure
The GOOSE transmission interval due to the occurrence of GOOSE events was repeated
30 times by the verification system simulator above, the passage time of the each
GOOSE at the point of the network analyzer was measured to the 4th transmission (GOOSE
intervals: Event-2-4-8ms). By calculating the difference between the measured transmission
times of the same packet, the movement time of GOOSE was calculated in the network.
And from the result, minimum, maximum and average value were determined as follows.
① network analyzer # 2 - # 1 network analyzer
: the delay measurement of the network switch inside connected to T/L IED
② network analyzer # 3 - # 2 network analyzer
: the delay measurement among network switches of “GOZAN”S/S
③ network analyzer # 3 - # 1 network analyzer
: the delay measurement from the T/L IED to the HMI
Table 1. Verification system for GOOSE subscription failure analysis
|
①
|
②
|
③
|
Max(sec)
|
0.000175
|
0.004911
|
0.004953
|
Min(sec)
|
0.000001
|
0.000001
|
0.0
|
Ave(sec)
|
0.000043
|
0.000098
|
0.000071
|
Considering that the communication traffic transmission performance of the network
system of “GOZAN” S/S is guaranteed in ‘ms’, over the average transmission time measured
in Table 1, within the interval of 2ms as the first transmission (SeqNum = 1) after
GOOSE event, the network transmission delay seems to be no problem. But, from the
maximum transmission time, transmission delay between the network switch is going
to approximately 5ms and the value is beyond 4ms, 2 times of the TAL. It does not
satisfy the conditions of IED receiving the GOOSE and BUS IED can cause the failure
notification of GOOSE subscription.
Fig. 9 shows one of transmission delay occurred in network system of the substation. Due
to 3.3ms of switch delay, BUS IED received second GOOSE message out of TAL 2ms of
first event GOOSE. “GOZAN” S/S was analyzed to have a failure of GOOSE subscription
from IEDs such as BUS IED because of the transmission delay of GOOSE that occurred
intermittently in the network system.
In order to solve this issue, when T/L IED generates second GOOSE, TAL value of the
GOOSE was changed to 10ms from 2ms. It is possible for all IEDs to process GOOSE subscription
from T/L IED enough regardless of the transmission delay measured above. Because the
present KEPCO policy on GOOSE application is limited to not breaker trip but interlocking
signals, this modification considering the delay time does not affect in the operation
of the digital substation. In the future, we think that it is necessary to improve
performance of the network system for process bus or tripping application of GOOSE.
Fig. 9. GOOSE interval is increasing more than 2ms in the network switch of “GOZAN”S/S
4. Conclusion
When Power utilities introduce the IEC 61850 based SAS to the new system of the substation,
the network system has been thought that there is no big problem. Therefore, we usually
find the cause in the devices that make up the digital substation such as IED or HMI
when a problem occurs in the digital substation. However, in this study, the network
performance of the digital substation sometimes can affect the operation of the substation
and a need for a method that can verify and correct them, was confirmed. In particular,
because GOOSE features a fast, repetitive transmission unlike MMS, verification system
and solution presented by this paper are expected to be very useful in positions of
power utilities that want to introduce digital substation. Through the results of
this on-site lesson, KEPCO plans to develop a device that can constantly monitor network
traffic such as GOOSE and communication systems of the digital substation and install
it on-site to ensure reliable operation of digital substation.
References
Skeie T., Johanessen S., Brunner C., Jun 2002, Ethernet in substation automation,
IEEE Control Systems, Vol. 22, No. 3, pp. 43-51
Christoph Brunner, 2012, Significant work is currently being done to further improve
the engineering capability and the interoperability level, PacWorld, Vol. 22, pp.
20-25
KEPCO , 2010, Power IT project final report: The Verification of Performance of the
Substation Automation System and Its Field test
Lee Nam Ho, Jang Byung Tae, Han Jeong Yeol, Lee You Jin, Choi Jong Kee, An Young Ho,
2014, Development of the IEC 61850 Server Simulator in order to Test a Digital Substation,
ICEE
IEC 61850-7-1 , 2003, Communication network and systems in substation- Part7-1: Basic
communication structure for substation and feeder equipment-Principles and models
IEC 61850-7-2 , 2005, Communication network and systems in substation- Part7-2: Basic
communication structure for substation and feeder equipment-Abstract communication
service interface (ACSI)
IEC 61850-8-1 , 2005, Communication network and systems in substation- Part-8-1: Specific
communication service mapping (SCSM) -Mapping to MMS (ISO/IEC 9506-1 and ISO/IEC9506-2)
and to ISO/IEC 8802-3
IEC 61850-10 , 2012, Communication network and systems for power utility automation
Part-10: Conformance testing
Biography
He was born in Korea in 1973. He received his B.S., M.S. and Ph.D. degrees in Electrical
Engineering from Myoungji University, Korea, in 1998, 2001, and 2011, respectively.
From 2004 to 2006, he was a junior researcher at LS Industrial Systems. He has worked
at KEPCO Research Institute since 2006 and is currently a senior researcher.
His research interests include IEC 61850, SAS, and Power restoration. He is a member
of the KIEE, KIIEE and IEC TC57 WG10.
He was born in Korea in 1988. He received his B.S. and M.S. degrees in Electrical
Engineering from Incheon National University, Korea, in 2013 and 2015.
He has worked at KEPCO Research Institute since 2015 and is currently a senior researcher.
His research interests include IEC 61850, SA, and RTDS. He is a member of the KIEE,
KIIEE and IEC TC57 WG10.
He was born in Korea in 1989. He received his B.S. and M.S. degrees in Electrical
Engineering from Gangneung-Wonju National University, Wonju, Korea, in 2014 and 2016.
He has worked on his Ph.D. in the Department of Electrical Engineering at Gangneung-Wonju
National University since 2016.
From 2015 to 2018, he was a junior researcher at ENTEC. He has worked at KEPCO Research
Institute since 2019 and is currently a researcher.
His research interests include SAS, IED, and Protection & Control. He is a member
of the KIEE and KIIEE.