LINE PROTECTION TESTS USING PROCESS BUS IEC 61850-9-2 WITH NETWORK LOADING

Summary

This paper reports real tests results that were obtained with a commercial IED that alreadyworks using the Process Bus, to verify the performance.

Firstly the IED was tested by using just one MU (Merging Unit) providing samples of Currents and Voltages to the IED through the LAN (Local Area Network). These samples were used by IED to measure the line impedance in order to decide about to trip or not. After exhaustive tests made with this arrangement the number of Merging Units connected to the LAN was increased, and the tests were repeated to investigate the effects of the loading to the system behavior (Up 10 MU ?s was connected together). The faults were carried on using a relay test set, and the trip time was measured for each case condition.

To evaluate the Process Bus behavior it were realized more than one thousand faults. The LAN was gradually loaded by increasing the number of MU connected to it. It will be evaluated if the trip time was affected with the increase in number of merging units.

The tests results presented are organized in tables and histograms formats, used to present the summary of the statistical analyses. This paper brings practical information for professionals from areas of Protection, Automation and Control interested in this new technology.

1. Introduction

The IEC 61850 is each day more present in new Substations and retrofits of old substations all over the world.

This standard basically addresses three different types of sending / receiving information: Client-Server, GOOSE (Generic Object Oriented Substation Events) and Sampled Value (SV).

The implementation of Report Communication (Client-Server), by MMS and the exchange of high speed GOOSE message are consecrated and widely used. However the digitalized current and voltage travelling on the Ethernet network called Sampled Value is not much explored yet.

The great difference between the high speed messages such as GOOSE and SV and the not high speed messages like reports is that the first works directly on the second layer, the link layer, so are faster than the messages that use the transport / network layers like Client – Server with MMS (Manufacturing Message Specification).

This application of Process Bus fascinates ones and frightens other member of our community, just because this standard topic is a paradigm shift from the traditional method of using secondary current and voltage obtained from instruments transformers that has been used for decades.

In order to make the Sampled Value a reality it is necessary using non-conventional CT ?s and VT ?s and / or Merging Units (MU) or SAMUs (Stand Alone Merging Units). Depending on the technology applied exists different ways to do it. These different forms are treated on the new standard IEC 61869 that is a complementary standard of the IEC 61850-9-2.

This change consists of working with digitalized signal of current and voltages travelling on an Ethernet network and distributing these messages to IED ?s. Therefore the IED, already receives the signal in digitalized (bits) format instead of doing the task of transforming the analog signal to digital inside it. On this way the A/D convertor was moved from the IED input to near the process.

By definition the Sampled Value Frame, has 3 Phase Current + Neutral Current also 3 Phase to Ground Voltages + Neutral Voltage, so we came to the reality that a substation will have a lot of Merging Units working in the Process Bus. Thereby the assay was made with many MUs connected on the network

There isn’t a global consensus about how to implement the SV, and it occurs because the IEC 61850 standard has left open some definition. So different proposes have emerged and the Light Edition (9-2 LE) has become the most popular.

The Light Edition was defined on the Implementation Guide of the UCA group however not everyone has adopted it because it ?s not a standard yet, although, it will be when the IEC 61869 becomes approved, since it standardizes the definitions presented in the Light Edition.

With adoption of Process Bus it starts a new era since it is a paradigm shift, because it changes for Ethernet network the current and voltage from the secondary hard wire that we have used for decades.

This new technology brings apprehension to some users. The traffic of information on the LAN (Local Area Network) causes fear due to the delays and losses that can lead to undue trip or even not trip on a fault condition.

So this paper intent to explorer deeply the Process Bus, by adopting different scenario conditions of the network loading (several MU sending samples values) and performing the tests to analyses if the IED behavior becomes affected.

2. The Tests

The tested object was a Line Protection IED. Many shoot tests were done to ascertain the impedance characteristic comparing to the characteristic defined by the setting.

The test ambient is composed of: Alstom IED P444, Relay Test Set CE-6006 Conprove with the SV accessory, Satellite Synchronization Unit CE-GPS Conprove, RuggedCom Switch and a Protocol Analyzer, illustrated in FIGURE 2.

Alstom GRID IED model MiCOM P444 works with the Sampled Value rate of 80 samples per cycle. The communication can be by Fiber Optic or Twisted Pair (RJ45). Also it already can work with the redundancy protocol PRP, but it is not the goal of this paper.

The IED under test was set to protect a line with 12 ohms and angle of 70 degrees, the Zones are: Z1 t=0s, Z2 t=200ms, Z3 t=1s (reverse), Z4 t=600ms.

Tests were executed with software named “Distanc”. It permits realize both shot tests and search tests Zones of Impedance Line Protection (21 / PDIS), the setting can be inserted in the software by importing RIO files or using masks. The applicative already allows entering the parameters of the mains relays in the market, changing the software screen as the relay specific set and nomenclature, making an easy work to the user. (FIGURE 3)

After the set was introduced, some search tests were made to confirm the setting, and one sample of the test is illustrated in FIGURE 4.

The test set provides fault conditions to the IED by sending the SV message carrying the instantaneous value of current and voltage. The IED sends to the test tool a TRIP signal using a cooper hard wire, that interconnects the IED binary output to the test set binary input. This test equipment has the capacity to simulate up to 12 MU ?s.

Tests Conditions: It were performed shot tests at many locations of impedance plane targeting to test all the 4 Zones parameterized further the external region that not operate, therefore 5 different regions of operation. The tests were done covering many zone ?s reach.

Type of Fault: The tests was realized with the three fault types LE, LL and LLL (that selected AE, AB and ABC), each fault was injected 20 times to do a statistical analyses.

Ethernet network load: the test scenario was changed to each load condition.

Initially the test was realized on a free network, i.e. the only traffic on the net was the MU that was providing samples to the IED. So it only had to capture these samples of current and voltage to calculate the impedance.

Many ways are possible to detect the message ?s sender, and normally the IED filters the field SV ID of the frame.

After were inserted others MUs on the network, which ones the IED does not have to measure the impedance. This condition will increase the network traffic. On each test it was added 3 new MUs. Therefore the second test with 4 MUs, the third with 7 e finally the fourth with 10 MUs. (FIGURE 5)

Totaling it was 20 x 5 x 3 x 4 = 1200 tests (repetitions x zones x fault types x Number of MUs)

It is important to mention that the entire test was carried on the line angle.

2.1 Analysis of the Results

Posteriorly the test phase been finished, it started the results analysis and statistics phase.

Firstly, for each test case (that were repeated 20 times), a table was made to point out the operation time, by showing maximum, minimum, average time and the standard deviation. So it was created 4 tables, one for each different network load condition. The TABLE 1, below illustrated the case of 10 MUs on the network.

Next, the analyses were made by type of fault. For each type of fault was analyzed the performance under different kinds of Ethernet loading, in order to verify if exist any relationship with trip delays and the load.

The test data was organized in chart forms seeking make an easy reading. (CHART 1)

As a final step it was summarized a final chart comparing the average trip time measured with 1 MU (net without load) and with 10 MUs on the network for all zones conditions.

The CHART 2 below shows the difference of average trip time of the two most divergent conditions. Showing the balance of time difference: (Average Trip Time with 10MUs – Average Trip Time with 1 MU).

3. Conclusion

The SV is regarded the future technology coming to substitute the secondary cooper hard wire from the Instrument Transformers. In the coming years it can be more used in new Substations or at Retrofits.

At this job the system was extensively tested with more than 1000 tests, and the results were analyzed, concluding that no pattern was observed comparing trip time delay and the increase of MU ?s on the network .Therefore, no relationship was established.

However, nevertheless, on the worst case if is compared the free network (just 1 MU) and the network with 10 MUs the worse average difference is 1.38 ms, in other words, less than a tenth cycle, a few significant or insignificant delay on operation time.

This job open the way for others test that can be realized in the future to complete this analysis and reinforce the concept of IEC 61850 standard Process Bus.

Bibliography

[1] Standard IEC 61850– Communication networks and systems in substation –– All Parts

[2] Implementa??o de um Sistema de Envio e Recebimento de Sampled Value (IEC 61850-9-2) - Pereira Junior, P. S,; Ramos, R. R.; Martins, C. M.; Pereira, P. S.; Louren?o, G. E.-Seminário Nacional de Produ??o e Transmiss?o de Energia Elétrica, XXI SNPTEE de 2011.

[3] Draft of the Standard IEC 61869-9 -Digital Interface for Instruments Transformers – August 2011

[4] Investiga??o do Funcionamento do Barramento de Processos (IEC 61850-9-2) - Uma Abordagem Prática - Pereira Junior, P. S,; Ramos, R. R.; Martins, C. M.; Pereira, P. S.; Louren?o, G. E.- Simpósio de Automa??o de Sistemas Elétricos, IX SIMPASE de 2011.

[5] Implementation Guideline for Digital Interface to Instrument Transformers Using IEC 61850-9-2 – UCA International Users Group – Dated 01/03/2004.

[6] IEC 61850-9-2 Avalia??o e Testes De Um Barramento De Processos - Pereira Junior, P. S,; Ramos, R. R.; Martins, C. M.; Pereira, P. S.; Louren?o, G. E. - Seminário Técnico de Prote??o e Controle, X STPC de 2010.

[7] Uma abordagem intensa do Barramento de Processos (IEC 61850-9-2), as inova??es da segunda edi??o e a rela??o com a norma de TC ?s e TP ?s – IEC 61869-9 - Pereira Junior, P. S,; Ramos, R. R.; Martins, C. M.; Pereira, P. S.; Louren?o, G. E. - Seminário Técnico de Prote??o e Controle, XI STPC de 2012.

[8] Experimento de Sobrecarga Com 15.000 Mensagens GOOSE por segundo em Uma Rede IEC 61850, e a Investiga??o de suas Consequências - Pereira Junior, P. S; Martins, C. M.; Pereira, P. S.; Louren?o, G. E.- Seminário Nacional de Produ??o e Transmiss?o de Energia Elétrica, XX SNPTEE de 2009.

[9] Testes de performance em IEDs através de ensaios utilizando mensagens GOOSE (IEC 61850) - Pereira Junior, P. S,; Martins, C. M.; Pereira, P. S.; Louren?o, G. E. - Seminário Técnico de Prote??o e Controle, IX STPC de 2008.

[10] Performance assessment of an IEC 61850-9-2 based protection scheme for the mesh corner of a 400KV transmission substation – P.A. Crossley, L. Yang, H.Y. LI, U. Anombem, A. Wen R. Chatfield, J. Wright, M. Redfern, X. SUN – CIGRE 2012