Historical Data

Getting Started

Setup Servers

Security

Authentication Security Access-control

Address Space

Address-space Data-types Methods Dynamic-variables Structures Extension-objects Views

Advanced

Events-and-alarms Historical-data

Integration

Testing-guide Ci-integration Customization

Historical Data

Path: Objects > TestServer > Historical

4 variables with historical data access enabled. The server records value changes in memory (up to 10,000 samples per variable) and serves them via the OPC UA HistoryRead service.

Historical Variables

BrowseName DataType Access Update Rate Value Pattern
HistoricalTemperature Double R + HR 1 second 22 + 8 * sin(t/60) + noise -- simulated temperature oscillation
HistoricalPressure Double R + HR 1 second 101.325 + 5 * cos(t/120) + noise -- simulated pressure
HistoricalCounter UInt32 R + HR 1 second Increments by 1 every second
HistoricalBoolean Boolean R + HR 1 second Random true/false every second

Recording Interval

Historical data is recorded at 1000ms (1 second) intervals.

Access Levels

These variables have:

  • accessLevel = CurrentRead | HistoryRead
  • userAccessLevel = CurrentRead | HistoryRead

The HistoryRead flag indicates that historical data is available.

Storage

  • Maximum samples: 10,000 per variable (circular buffer -- oldest samples are discarded)
  • Storage type: In-memory (not persisted across restarts)
  • Sample rate: 1 sample/second for all variables (1000ms recording interval)

After the server has been running for ~2.8 hours, the buffer is full and oldest samples start being overwritten.

Supported HistoryRead Operations

ReadRawModifiedDetails

Read raw historical values within a time range.

Parameters:

  • StartTime -- beginning of the time range
  • EndTime -- end of the time range
  • NumValuesPerNode -- maximum number of values to return (0 = no limit)
  • ReturnBounds -- whether to include bounding values
  • IsReadModified -- read modified values (not applicable here)

Example request:

StartTime: 5 minutes ago
EndTime: now
NumValuesPerNode: 100

Returns up to 100 raw samples from the last 5 minutes, each with timestamp and value.

ReadProcessedDetails

Read aggregated historical values (e.g., average, min, max over intervals).

Parameters:

  • StartTime / EndTime -- time range
  • ProcessingInterval -- aggregation interval in milliseconds
  • AggregateType -- NodeId of the aggregate function

Common aggregates:

  • Average
  • Minimum
  • Maximum
  • Count

ReadAtTimeDetails

Read the interpolated value at specific timestamps.

Parameters:

  • ReqTimes -- array of timestamps to read values at
  • UseSimpleBounds -- interpolation method

Testing Historical Data

Basic history read

  1. Connect to the server and wait at least 10 seconds (so history accumulates)
  2. Read HistoricalTemperature with ReadRawModifiedDetails:
    • StartTime = 10 seconds ago
    • EndTime = now
  3. Expect ~10 samples, each approximately 1 second apart
  4. Values should be around 22 with +/-8 oscillation

Pagination

  1. Set NumValuesPerNode = 5
  2. Read history -- should return 5 values plus a ContinuationPoint
  3. Call HistoryRead again with the ContinuationPoint to get the next batch
  4. Repeat until ContinuationPoint is null

Time range filtering

  1. Read HistoricalCounter from startup to now
  2. Values should be sequential: 0, 1, 2, 3, ...
  3. Read a sub-range and verify the values match the expected range

Boolean history

  1. Read HistoricalBoolean history
  2. Values should be a random sequence of true/false
  3. Each sample should have a unique timestamp ~1 second apart

Aggregates (if supported by your client)

  1. Read HistoricalTemperature with ReadProcessedDetails:
    • ProcessingInterval = 60000 (1 minute)
    • AggregateType = Average
  2. Each returned value should be the average temperature over that minute
  3. Compare with Minimum and Maximum aggregates