Repositories
All STEPSS repositories are hosted under the SPS-L GitHub organization.
Repository Index
Section titled “Repository Index”| Repository | Role | Language | Visibility |
|---|---|---|---|
| stepss-java-ui | Java-based GUI front-end, bundling the whole toolchain | Java | Public |
| stepss-python-ui | Python API/wrapper for RAMSES and the Helios power-flow engine | Python | Public |
| stepss-userguide | LaTeX user documentation and models reference | LaTeX | Public |
| stepss-uramses | User-defined device models framework | Fortran | Public |
| stepss-eigenanalysis | Reference data and validation suite for small-signal analysis | Python | Public |
| stepss-cg-studio | Visual block diagram editor for CODEGEN models | Python/JS | Public |
| stepss-ramses | Core RAMSES simulation engine | Fortran | Private |
| stepss-test-systems | Curated collection of test cases and network models | RAMSES data | Private |
| stepss-helios | AC power-flow engine (Newton-Raphson), with a C API shared library wrapped by stepss (stepss.helios); releases ship CLI and C API binaries for Linux/macOS/Windows | C++ | Private |
| stepss-dyngraph | Dynamic graph / topology module | Fortran | Private |
| stepss-Codegen | DSL-to-Fortran model code generator | Fortran | Private |
| stepss-RamsesNN | Physics-informed neural network experiments on RAMSES models | Python | Public |
| stepss-docs | This documentation website | Astro/Starlight | Public |
Public Repositories
Section titled “Public Repositories”STEPSS (GUI)
Section titled “STEPSS (GUI)”The main graphical user interface for STEPSS, built with Java (Swing/AWT) and the Ant build system.
- Repository: github.com/SPS-L/stepss-java-ui
- License: Apache License 2.0
- Requirements: 64-bit Java 11 or later to run the jar, none at all to run an installer, JDK plus Apache Ant to build
- Build:
ant jar - Run:
java -jar dist/stepss.jar, orant bundlefor a native installer
stepss-python-ui
Section titled “stepss-python-ui”Python interface to the RAMSES simulator providing scripting access to simulations, plus the stepss.helios module for AC power flows with the bundled Helios engine (Windows, Linux, and macOS).
- Repository: github.com/SPS-L/stepss-python-ui
- Install:
pip install stepss - Documentation: Python API section on this site
- Includes: five runnable power-flow examples under
examples/helios/
stepss-userguide
Section titled “stepss-userguide”The original LaTeX source for the STEPSS documentation (models reference and user guide).
- Repository: github.com/SPS-L/stepss-userguide
- Build:
pdflatex stepss_doc.tex(run twice) - Main file:
stepss_doc.tex
stepss-uramses
Section titled “stepss-uramses”Framework for compiling and linking custom Fortran models with RAMSES.
- Repository: github.com/SPS-L/stepss-uramses
- Platforms: Linux (gfortran), macOS (gfortran, Apple Silicon), Windows (MinGW gfortran or Intel Fortran)
- Build (Linux):
make -f build/Makefile.linux all - Build (macOS):
make -f build/Makefile.macos all - Build (Windows/MinGW):
make -f build/Makefile.windows all - Build (Windows/Intel): Visual Studio with
build/msvs/URAMSES.sln
stepss-eigenanalysis
Section titled “stepss-eigenanalysis”Reference spectra and the validation suite for RAMSES’s built-in small-signal analysis. The analysis itself runs in the engine, documented under Eigenanalysis; this repository holds the independently captured reference data the engine is checked against, so its tests need neither a RAMSES licence nor the engine itself.
- Repository: github.com/SPS-L/stepss-eigenanalysis
- Requirements: Python with numpy and pytest
stepss-cg-studio (CODEGEN Studio)
Section titled “stepss-cg-studio (CODEGEN Studio)”Browser-based visual editor for building CODEGEN user-defined models with drag-and-drop blocks.
- Repository: github.com/SPS-L/stepss-cg-studio
- Requirements: Python 3.10 or later. On macOS also
brew install gcc, which the bundled CODEGEN links against. - Install:
pip install stepss-cg-studio - Run:
cg-studio→ openhttp://localhost:8765 - Documentation: CODEGEN Studio guide on this site
The wheel bundles the CODEGEN executables for Linux, Windows and macOS, so Run Codegen works on a fresh install with nothing to configure. That makes it mixed-licence: see License. Its version names the CODEGEN it carries, so 5.3.0 and 5.3.1 both run CODEGEN 5.3.