Skip to content

Releases: dnv-opensource/component-model

v0.3.1

17 Dec 09:50

Choose a tag to compare

What's Changed

  • Address breaking changes with PythonFMU's latest changes to setup_experiment() function.

v0.3.0: ECCO Support

15 Dec 08:47

Choose a tag to compare

Added

  • Added the documentation file component-development-process.rst, outlining the recommended FMU development workflow, best practices, and the role of virtual derivatives.
  • Added the example XML structures BouncingBallStructure.xml and ForcedOscillator6D.xml to showcase richer algorithm and ECCO configuration sections together with clarified causality definitions.

Changed

  • Refined the existing FMU XML examples with explicit algorithm sections, ECCO configuration, and clearer variable linkage descriptions.
  • Updated driving_force_fmu.py to support vectorized amplitudes, frequencies, and frequency sweeps, while improving type annotations and initialization semantics.
  • Changed the upper bound for the height variable in bouncing_ball_3d.py to inches to align with the rest of the example unit system.

GitHub workflows

  • Added Python 3.13 to the main CI matrix to ensure compatibility with the current stable release.
  • Updated the future/experimental workflow to track Python 3.14 instead of 3.13 for forward-looking coverage.

v0.2.0

30 Apr 11:25

Choose a tag to compare

What's New: Structured Variables

Introducing structured variables to component-model according to FMI standard v.2 and 3.

Key Benefits:

  • Hierarchical Variable Organization: Variables can now be organized in a tree structure using dot notation (e.g., vehicle.engine.speed)
  • Automatic Derivative Handling: Define interface variables that represent derivatives (like speeds) of base variables without explicitly defining them in the underlying model
  • Simplified Complex System Modeling: Models can expose a more intuitive interface while abstracting internal complexity

Technical Improvements:

  • New variable naming conventions: flat and structured
  • Support for derivative notation: der(variable) and der(variable,n)
  • Enhanced parent-child relationship handling for nested components
  • Improved array indexing with structured notation

v0.1.0

08 Nov 11:41
2ee3760

Choose a tag to compare

Changed

  • Changed from pip/tox to uv as package manager
  • README.rst : Completely rewrote section "Development Setup", introducing uv as package manager.
  • Changed publishing workflow to use OpenID Connect (Trusted Publisher Management) when publishing to PyPI

GitHub workflows

  • (all workflows): Adapted to use uv as package manager
  • _test_future.yml : updated Python version to 3.13.0-alpha - 3.13.0
  • _test_future.yml : updated name of test job to 'test313'

v0.0.2b3

08 Nov 10:12
a1c0896

Choose a tag to compare

v0.0.2b3 Pre-release
Pre-release

GitHub workflow publish_release.yml Test no. 3

v0.0.2b2

08 Nov 09:46
8160734

Choose a tag to compare

v0.0.2b2 Pre-release
Pre-release

GitHub workflow _publish_package.yml Test no. 2

v0.0.2b1

05 Nov 15:04

Choose a tag to compare

v0.0.2b1 Pre-release
Pre-release

GitHub workflow _publish_package.yml Test no. 1

First release

27 Sep 08:38

Choose a tag to compare

This is the first release of the Component Model package, designed to extend the capabilities of the PythonFMU framework. The package allows for the seamless creation and simulation of component models adhering to the FMI, OSP, and DNV-RP-0513 standards. It focuses on efficient model-to-FMU translation and supports vector-based variable manipulation, unit definitions, and range checking. The package also enhances the Assurance of Simulation Models, with a particular focus on DNV-RP-0513 compliance.

Key Features:

Model-to-FMU Conversion:
Efficiently convert Python-based models to FMU (Functional Mock-up Units) with minimal code overhead.

  • Vector Variable Support: Seamless handling of vector variables using numpy, enabling 3D model simulations and unit-based calculations.

  • Unit and Display Unit Management: Full support for defining units and display units of variables, ensuring consistency in model representation.

  • Range Checking for Variables: Built-in support for validating variables against specified ranges, enhancing model integrity and reducing runtime errors.