Time Series Data

Time Series Data#

As a BaSyx AAS Web UI user
I want to visualize Time Series Data Submodels using different chart types
so that I can analyze temporal data patterns and trends.

Semantic ID#

This plugin is activated when a Submodel has the following semantic ID:

  • V1.1: https://admin-shell.io/idta/TimeSeries/1/1

Feature Overview#

The Time Series Data plugin provides powerful visualization capabilities for time-based data stored in Asset Administration Shells according to the IDTA Time Series Data specification. It supports three segment types for storing and accessing time series data:

  1. InternalSegment: Time series data stored directly within the AAS Submodel

  2. ExternalSegment: Time series data stored in files or blob SubmodelElements (e.g., CSV files)

  3. LinkedSegment: Time series data from linked time series databases (e.g., InfluxDB)

The plugin offers multiple chart types for visualization, making it ideal for monitoring, analysis, and historical data review.

Note

For a complete working example with InfluxDB, MQTT, and all three segment types, see the TimeSeriesData example on GitHub.

Time Series Data Plugin

Time Series Data Plugin#

Key Features#

  • Three Segment Types:

    • InternalSegment: Data defined within the AAS

    • ExternalSegment: Data from files or blob SubmodelElements (CSV with RFC 4180 format)

    • LinkedSegment: Real-time data from time series databases (InfluxDB tested)

  • Multiple Chart Types: Line Chart, Area Chart, Scatter Chart, Histogram, Gauge, Display Field

  • Unified Time Range: One relative or absolute range controls both data retrieval and every visualization

  • Optional Auto Refresh: Re-fetch a valid configuration at an interval in seconds, minutes, or hours

  • Chart Configuration Options:

    • Interpolation mode

    • Number of bins (for Histogram)

    • Stacked bars (for Histogram)

  • Interactive Charts: Zoom, pan, and hover for detailed information

  • Multi-series Support: Display and compare multiple y-variables simultaneously

  • Database Integration: Direct connection to InfluxDB with API token authentication

  • Variable Selection: Choose time variable and multiple y-variables from the submodel metadata

Usage#

General Steps#

  1. Navigate to a Submodel with the Time Series Data semantic ID in the AAS Treeview

  2. Open the Visualization tab

  3. In the Preview Configuration card:

    • Select a Segment (Internal, External, or Linked)

    • Select the time variable from the metadata records

    • Select one or more y-variables to visualize

    • Use the adjacent Time Range and Auto Refresh menus in the top-right corner

  4. Click Fetch Data to resolve and commit the range, then load matching data

  5. In the Preview Chart card:

    • Select a Chart Type (Line, Area, Scatter, Histogram, Gauge, Display Field)

    • Adjust chart-specific options such as interpolation or histogram bins

    • View the visualization

Time Range#

The compact Time Range menu in the top-right of Preview Configuration applies to the query and to all chart types. Its button always shows the active range. It defaults to the last minute and is session-local: reopening the plugin restores the default.

Relative ranges can use a preset (1m, 5m, 15m, 1h, 6h, 12h, 24h, 7d, 30d, or 1y) or a custom positive value in milliseconds, seconds, minutes, hours, days, weeks, months, or years. Months and years use calendar-aware subtraction.

Absolute ranges use local start and end date-times. The plugin converts them to UTC before querying and rejects incomplete or reversed ranges.

When Fetch Data is clicked, the relative range is resolved once:

  • LinkedSegment ranges end at the current time.

  • InternalSegment and ExternalSegment ranges end at the newest available timestamp, which keeps archived data usable.

  • Absolute ranges always use their exact start and end.

Changing the selector does not alter LinkedSegment StartTime or EndTime; those fields describe the segment data rather than transient visualization state. If the committed range contains no records, the plugin shows an empty state instead of stale chart data.

Auto refresh#

The Auto Refresh menu sits directly beside the Time Range menu and shows Off or the active interval on its button. Auto refresh is disabled by default with an initial interval of 30 seconds. Enable it and enter a positive interval in seconds, minutes, or hours. Refreshing starts after a segment, time variable, y-variable, and valid range are selected. A tick is skipped while an earlier request is still running.

Each refresh resolves a relative LinkedSegment range against the current time. Absolute ranges keep their exact bounds. InternalSegment relative ranges remain anchored to the newest available record, while ExternalSegment data is requested again and anchored to the newest timestamp in the returned file. Auto refresh preserves a manual chart zoom; fetching a changed time range resets it.

Segment Types#

InternalSegment#

InternalSegment defines time series data directly within the AAS Submodel structure. Data points are stored as properties with timestamps and values, making it suitable for small to medium-sized datasets that don’t change frequently.

Characteristics:

  • Data embedded in the Submodel

  • No external dependencies

  • Best for static or infrequently updated data

  • Immediate availability

ExternalSegment#

ExternalSegment references time series data stored in files or blob SubmodelElements. The plugin supports CSV files following RFC 4180 format with a header line.

Characteristics:

  • Data stored in separate files

  • Supports larger datasets

  • CSV format with headers (time column + value columns)

  • Referenced via File SubmodelElement

  • Suitable for archived or batch data

LinkedSegment#

LinkedSegment connects to external time series databases for real-time or large-scale data. The example includes InfluxDB with Telegraf for MQTT data ingestion.

Characteristics:

  • Real-time data access

  • Scalable for large datasets

  • Supports time series databases (InfluxDB tested)

  • Configurable queries using Flux language

  • API token authentication

  • Supports standard Flux time variables and the {{y-value}} field placeholder

Database Configuration:

  • Endpoint: URL of the time series database

  • Query: Flux query to fetch data (supports the variables documented below)

  • API Token: Authentication token entered in the UI or supplied as INFLUXDB_TOKEN to the Web UI container. Source/development builds can instead set the build-time VITE_INFLUXDB_TOKEN variable.

Flux query variables#

Flux queries copied from the InfluxDB query editor or a Grafana Flux panel can use these variables without modification:

Variable

Value supplied by the plugin

v.timeRangeStart

Selected range start as a UTC RFC 3339 time literal

v.timeRangeStop

Selected range end as a UTC RFC 3339 time literal

v.windowPeriod

Range duration divided into approximately 360 points, rounded to milliseconds with a minimum of 1ms

The following query is ready to paste into a LinkedSegment Query property:

from(bucket: "basyx")
  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)
  |> filter(fn: (r) => r["_measurement"] == "machine_metric")
  |> filter(fn: (r) => r["_field"] == "{{y-value}}")
  |> aggregateWindow(every: v.windowPeriod, fn: mean, createEmpty: false)
  |> yield(name: "mean")

The three v.* variables describe the committed time range. The separate {{y-value}} placeholder is replaced with the y-variable selected in the plugin. In the linked segment from the working example, select rpm, pressure, or vibration with the query above. Other v.* references are left untouched, so variables defined inside the Flux query continue to work; unresolved database variables are reported by InfluxDB. The plugin does not rewrite hard-coded range() calls or provide arbitrary Grafana dashboard variables.

Supported Chart Types#

Line Chart#

Best for continuous data and trend analysis. Displays time series as connected lines.

Configurable options:

  • Interpolation mode

Area Chart#

Emphasizes volume and cumulative values by filling the area under the line.

Configurable options:

  • Interpolation mode

Scatter Chart#

Shows individual data points without connecting lines, ideal for identifying patterns and outliers.

Line, Area, and Scatter charts use the time range committed in Preview Configuration. Their toolbar provides x-axis selection, zoom in/out, pan, and reset controls, hovered values include their configured units, and the y-axis scales to the visible data. Scrolling the page while the pointer is over a chart does not zoom it; zooming requires an explicit chart action. Auto refresh preserves a manual zoom so an inspected section remains stable. Fetching a changed time range resets the viewport, and the chart reset control returns to the latest committed range.

Histogram#

Displays the distribution of values in bins, useful for frequency analysis.

Configurable options:

  • Number of bins

  • Stacked bars option

Gauge#

Displays the latest value as a gauge, ideal for real-time monitoring of current states. Its value, label, and unit remain visible without hovering. When multiple y-variables are selected, each valid series is rendered as a separate responsive gauge.

Display Field#

Shows the most recent value for each y-variable in a simple text display with units. Numeric values are rounded to two decimal places.

Working Example#

A complete Docker Compose example demonstrating all three segment types is available in the BaSyx AAS Web UI repository:

TimeSeriesData Example

The example includes:

  • InfluxDB time series database

  • Telegraf for MQTT data ingestion

  • MQTT publisher for simulated sensor data

  • Pre-configured AAS with all three segment types

  • Demo data for immediate testing

References#