Skip to content

VALIDATION_TEST_PLAN.md

Latest commit

 

History

History
107 lines (70 loc) · 3.07 KB

VALIDATION_TEST_PLAN.md

File metadata and controls

107 lines (70 loc) · 3.07 KB
Raw

Validation Test Plan - Raw Passive SV/GOOSE Analyzer

Purpose

Provide a focused validation route for Process Bus Insight as a public engineering tool. The goal is not to certify a timing instrument. The goal is to verify that the analyzer is raw-passive, decodes process-bus frames internally, binds live traffic to SCL context, and reports stream/timing anomalies honestly.

Test 1 - Product Dependency Boundary

Expected result:

  • ProcessBus.App.Wpf references only ProcessBus.Core and ProcessBus.Iec61850.Raw.
  • No product project reference to vendor wrapper code or external IEC 61850 subscriber stacks.
  • Product output does not require iec61850.dll.

Evidence:

  • project file screenshot or source review,
  • README statement,
  • solution project list.

Test 2 - Adapter Selection Evidence

Procedure:

  1. Open app.
  2. Select physical Ethernet adapter.
  3. Start capture.
  4. Copy evidence.

Expected result:

  • evidence contains adapter display name and raw device name,
  • timing confidence is not LOW unless adapter is virtual/USB/wireless,
  • app warns clearly if adapter is unsuitable.

Test 3 - SV Stream Decode

Procedure:

  1. Feed IEC 61850-9-2LE style SV traffic from Omicron, a bench publisher, or a known process-bus source.
  2. Observe stream explorer.
  3. Select stream.

Expected result:

  • stream card shows svID, APPID, and RAW status,
  • diagnostics shows raw APDU decoded,
  • sequence/missing-sample counters are visible,
  • metering/phasor/scope update when mapping is available.

Test 4 - Arrival Timing Excursion Screening

Procedure:

  1. Use a controlled source capable of introducing delay/arrival variation, or use a known unstable capture path for screening demonstration.
  2. Run capture for at least 10 seconds.
  3. Open Diagnostics.
  4. Copy evidence.

Expected result:

  • >=300 us event count increments when excursions occur,
  • event log records latest svID/APPID/smpCnt/delta/arrival variation,
  • interpretation differentiates clean smpCnt from capture timing path suspicion.

Acceptance wording:

  • Pass as anomaly screening if event is detected and capture confidence is stated.
  • Do not treat as measurement-grade proof unless externally validated.

Test 5 - Stale/No Stream State

Procedure:

  1. Start capture while stream is present.
  2. Stop source/publisher.
  3. Observe UI after stale timeout.

Expected result:

  • stream card becomes STALE,
  • diagnostics says no live stream or SV stale,
  • UI does not claim no traffic if packets were already observed.

Test 6 - GOOSE Passive Discovery

Procedure:

  1. Feed or observe GOOSE traffic.
  2. Open GOOSE tab.

Expected result:

  • GOOSE traffic appears as passive decoded events,
  • no control workflow is exposed,
  • state/event history is append-oriented and readable.

Final Acceptance Criteria

The product is ready for broader user evaluation when:

  • raw product boundary is clear,
  • no external IEC 61850 stack dependency appears in product solution,
  • timing claims are honest and confidence-graded,
  • event/evidence copy is repeatable,
  • physical Ethernet validation path is prepared,
  • known limitations are documented.