Test Systems / RBTS

Roy Billinton Test System (RBTS)

Original 1989 six-bus reliability test system, published distribution and overall-system extensions, source-attributed numerical tables, original-form configuration diagrams and contributed engineering cases.

Research release — verification required. Care has been taken to match the published IEEE test-system descriptions. Users MUST check data, equipment connectivity, operating assumptions and results against the cited source before relying on any file. Report omissions, revisions and discrepancies to sagarwal@gri-us.com. Revisions will be reviewed, attributed and versioned. Only available native file formats are provided.

Original reference and evolution

The 1989 RBTS comprises 6 buses, 11 generating units and 9 transmission lines. Its geographically suggestive diagram depicts lines in proportion to their lengths. The 1991 publication introduces Bus 2 and Bus 4 distribution systems; later work extends it toward overall-system assessment.

1989 Basic Data · 1990 Basic Results · 1991 Basic Distribution Data and Results

Numerical datasets

Selected available research tables. Numerical transcription remains under review. Original data and later published extensions are not merged silently.

YearDatasetFormats
1989Generation unitsCSV · JSON · TXT
1989Generator reliabilityCSV · JSON · TXT
1989Generator costsCSV · JSON · TXT
1989Bus peak loadsCSV · JSON · TXT
1989Power-flow dataCSV · JSON · TXT
1989Transmission electricalCSV · JSON · TXT
1989Transmission outage ratesCSV · JSON · TXT
1989Station componentsCSV · JSON · TXT
1989Load-duration data (review only)CSV · JSON · TXT
1989Customer interruption damageCSV · JSON · TXT
1991Customer load pointsCSV · JSON · TXT
1991Feeder loadingCSV · JSON · TXT
1991Feeder section lengthsCSV · JSON · TXT
1991Distribution assumptionsCSV · JSON · TXT
1991Published system indicesCSV · JSON · TXT

Original RBTS network and station-configuration diagrams

These five drawings were supplied by the model developer. They distinguish the basic network model from four alternative station arrangements. No abstract or conceptual distribution overview is presented as a source-equivalent SLD.

RBTS original six-bus network without detailed substations — original contributor-supplied diagram
RBTS original six-bus network without detailed substations
RBTS No Stations.jpg
Open / download original JPEG
Breaker-and-a-half arrangement — original contributor-supplied diagram
Breaker-and-a-half arrangement
RBTS Breaker One & One Half.jpg
Open / download original JPEG
Breaker one-and-one-third arrangement — original contributor-supplied diagram
Breaker one-and-one-third arrangement
RBTS Breaker One & One Third.jpg
Open / download original JPEG
Double-bus double-breaker (DBDB) — original contributor-supplied diagram
Double-bus double-breaker (DBDB)
RBTS DBDB.jpg
Open / download original JPEG
Ring-substation arrangement — original contributor-supplied diagram
Ring-substation arrangement
RBTS Ring.jpg
Open / download original JPEG
File publication status: The images are referenced at their intended permanent gri-us.com paths, but binary image upload is pending. The original JPEGs are preserved unchanged. The detailed Bus 2 and Bus 4 distribution single-line diagrams from the 1991 paper will be added only when accurate versions are available.

Engineering models and SUBREL study files

Original native files are organized by RBTS configuration and study purpose. Each listed file is a separately supplied engineering artifact; no unsupported conversion formats have been added. File publication status: all fifteen contributed engineering files are now hosted on the GR main website and individually downloadable. The original five station JPEG diagrams above are also available.

Basic RBTS — no detailed substations

Configuration figure: RBTS No Stations.jpg

Supplied filePurposePublication
RBTS.RAWPSS/E six-bus bus-branch caseDownload available
RBTS.savPSS/E saved binary caseDownload available
rbts.pwbPowerWorld caseDownload available
RBTS.pwdPowerWorld one-line drawingDownload available
rbts.AUXPowerWorld auxiliary case dataDownload available

The original six-bus representation is distinct from the detailed station models.

Breaker-and-a-half substations

Configuration figure: RBTS Breaker One & One Half.jpg

Supplied filePurposePublication
RBTSBreakerandHalf.pwbPowerWorld caseDownload available
RBTSBreakerandHalf.pwdPowerWorld one-line drawingDownload available

The original contributor-supplied PWB and PWD are a matching model/display pair by filename.

Ring substations

Configuration figure: RBTS Ring.jpg

Supplied filePurposePublication
RBTSRingSubstations.rawPSS/E RAW networkDownload available
RBTSRingSubstations.pwbPowerWorld caseDownload available
RBTSRingSubstations.pwdPowerWorld one-line drawingDownload available

The RAW file includes explicit additional switching nodes compared with the six-bus base case.

Ring-substation SUBREL study

Supplied filePurposePublication
RBTS Ring N-1F N-1S SUBREL.subrelSUBREL projectDownload available
RBTS Ring N-1F N-1S SUBREL.mdbMicrosoft Access databaseDownload available
RBTS Ring N-1F N-1S SUBRELPSSE.conPSS/E contingency exportDownload available
RBTS Ring N-1F N-1S SUBRELPWCon.auxPowerWorld contingency exportDownload available

These are related SUBREL project/database/export files; the 484-case contingency export scope must not be confused with the broader historical workbook.

Three-configuration historical reliability results

Supplied filePurposePublication
RBTS N-2F N-1S SUBREL Results.xlsxOriginal workbook: no detailed stations, breaker-and-a-half, ringDownload available

Historical SUBREL results pending independent InfraRel reproduction. Summary values are already available below.

Other station figures without matching model files in this collection

Breaker one-and-one-third and double-bus double-breaker (DBDB) are available as original JPEG diagrams. Corresponding PWB/PWD pairs have not been supplied and are not implied.

Verification: File availability is not evidence that case data or study outputs have been independently reproduced. Users must verify the connectivity, parameters, operating states, events and results against the applicable source before analysis, and report discrepancies to sagarwal@gri-us.com. Source files are preserved rather than automatically converted or corrected.

Historical SUBREL comparison: three RBTS substation arrangements

Historical study · Supplied workbook · Not independently reproduced

This worked example compares the original RBTS without detailed substations, a breaker-and-a-half arrangement, and a ring-substation arrangement. Results below are transcribed from the author's RBTS N-2F N-1S SUBREL Results.xlsx workbook. They are not results from InfraRel. The study scope includes up to N-2 forced faults and N-1 stuck-breaker conditions as described in the workbook, subject to configuration-specific enumeration.

Reported system reliability indices

IndicatorNo detailed substationsBreaker-and-a-halfRing
SAIFI (interruptions/customer-year)1.996912.719033.20337
SAIDI (minutes/customer-year)618.83838823.166871,051.01868
CAIDI (minutes/interruption)309.89731302.74243328.09814
ASAI0.99882262950.99843382840.9980003238
Energy unserved (kWh/year)259,547729,138826,127
Reported outage costs ($/year)1,213,54526,140,64427,725,088

Contingency enumeration

CategoryNo detailed stationsBreaker-and-a-halfRing
N-1 forced outage3112192
N-1 isolation2912192
N-2 forced outage4065,8823,721
N-1 isolation + N-1 forced outage83014,5208,372
N-1 forced outage + N-1 stuck0426300
Credible contingencies1,29621,07012,577
Contingencies causing partial load loss2265,5224,952
Interpretation caution. These results are not evidence that adding breaker-and-a-half or ring stations necessarily reduces reliability. They reflect the legacy study's particular models, fault definitions, outage rates, isolation and restoration assumptions, customer data and cost basis. The no-detailed-stations configuration represents a different modeling scope. The comparatively large cost differences require audit. The workbook's Project Settings tab identifies the no-detailed-station run while its summary sheets contain three project names; exact alignment of run settings is not established.

Supporting SUBREL files and planned reproducibility

A separate ring-substation SUBREL project and Microsoft Access database, with PSS/E and PowerWorld contingency exports, have been supplied and preserved. A 484-case ring N-1 fault / stuck-breaker export set is not equivalent to the complete 12,577-credible-contingency enumeration shown above; the study scopes must be reconciled before comparison. Original project/database and model binaries are being prepared for GR-domain distribution.

Reference results and other studies

Published 1991 distribution-system indices are included in the review datasets. Historical SUBREL three-configuration summary results are now shown above; original 1990 results and new InfraRel comparison outputs remain to be reproduced and source-checked. Renewal, storage, microgrid and AI variations will be published with separate provenance and assumptions.

Research history

The original system was developed by the University of Saskatchewan Power Systems Research Group under Professor Roy Billinton. Dr. Sudhir Kumar Agarwal, who published as S. Kumar, coordinated the 1988–1989 research and was the principal writer of the foundational papers. The repository is maintained as an open educational and technical resource.