hermess is a dynamic simulation tool for power systems modeled by nonlinear Differential-Algebraic Equations (DAEs).
It integrates the dynamic evolution equations of the components together with the algebraic network equations to produce time-domain trajectories of the bus voltages and of every device's internal states.
This repository is the simulation-only build of PowerDynamicEstimator: the dynamic state estimation (DSE) layer has been removed, leaving a focused, self-contained power-system dynamic simulator. The Python package is still importable as hermess.
- Time-domain simulation of nonlinear DAE power system models
- Synchronous machine models (transient and subtransient, including Sauer–Pai formulations) with pluggable AVR, governor, PSS and (multi-mass) shaft strategies
- Grid-forming and grid-following converter (inverter) models with composable control strategies
- Static loads (constant power / impedance / ZIP) and infinite bus
- Quasi-static or fully dynamic (differential) network/line models
- Disturbances: bus/line faults, line switching, and load steps
- Configurable reference-frame handling (centre-of-inertia, single-machine, nominal, distributed)
- Small-signal eigenvalue / participation-factor analysis at the operating point
- Multiple integration schemes (
idas,cvodes,collocation,rk) - Supports 50 Hz and 60 Hz systems
- User-defined dynamic and static models can be integrated
- Available as a Python package:
hermess
pip install hermessWith the optional desktop GUI:
pip install "hermess[gui]"- Clone the repository and
cdinto it. - Create and activate a virtual environment:
python -m venv venvsource venv/bin/activate # On Windows: venv\Scripts\activate- Install the package:
pip install -e .-
Clone the repository and
cdinto it. -
Create the
condaenvironment:
conda env create -f environment.yamlconda activate hermessThe package installs a hermess command (equivalently python -m hermess):
hermess list # the ready-made systems
hermess run 3bus --t-end 5 # simulate one and plot the trajectories
hermess run ieee39_conv --small-signal # the shipped demo scenariohermess run accepts --t-end, --ts, --no-plot, --small-signal and
--system-root for systems of your own; any other simulation setting is
reachable with --set KEY=VALUE (see hermess run --help).
From Python, to run a system without editing config.py:
import hermess
hermess.list_systems() # the ready-made systems
dae = hermess.simulate("3bus", T_end=5.0) # run one; returns the finished modelUser-written models are selectable from a system file once registered, and the same call takes devices and every pluggable strategy (AVR, governor, PSS, shaft, converter filter / angle / voltage / inner / PLL):
hermess.register(VSMAngle, "VSM") # now: angle = "VSM" in sim_param.txt
hermess.registered("angle") # what can be selected todayAn optional desktop GUI covers the interactive workflow: select a shipped or user-written system, adjust the simulation options, run with live progress, and inspect the one-line diagram, the time-domain trajectories, the small-signal modes and the initial power flow. Systems can also be built graphically, by placing buses, lines and devices on the canvas with parameter forms generated from the models; the result is saved as an ordinary system folder. Results export as plot-ready CSV.
pip install "hermess[gui]" # from source: pip install -e ".[gui]", or uv sync --extra gui
hermess-gui
See the GUI guide for a tour.
You can check out the available examples in the ./examples directory to get started, and the documentation for the full usage guide and API reference.
System dynamic and static parameters, including the topology, are specified in the ./hermess/systems subfolder. You can define the loads, generators, converters, and their characteristics at specific nodes in the power system. The component parameters live in sim_param.txt and the disturbances in sim_dist.txt within each system folder.
Adjust parameters related to the simulation (time step, end time, integration scheme, reference-frame mode, plotting, small-signal analysis, etc.) in the ./hermess/config.py file.
- Injector Limitation: Currently, the platform supports only one injector per node due to initialization ambiguity. To handle multiple injectors per node, you can create a new node connected via a branch with very small impedance.
© 2024-2026 ETH Zurich
Created by: Milos Katanic (original author of PowerDynamicEstimator) and Maitraya Avadhut Desai (simulation-only fork and maintainer).
HERMESS is a simulation-only fork of PowerDynamicEstimator (https://doi.org/10.5905/ethz-1007-842); the dynamic state estimation has been removed. See AUTHORS and CONTRIBUTORS for the full list of authors and contributors, and CONTRIBUTING.md if you would like to contribute.
PowerDynamicEstimator, from which this build is derived, was developed at the Power Systems Laboratory at ETH Zurich, supported as part of NCCR Automation, a National Centre of Competence in Research funded by the Swiss National Science Foundation (grant number 51NF40_225155).
If you use HERMESS in academic work, please cite the software release in the ETH Research Collection (see also CITATION.cff):
M. A. Desai, M. Katanic, and G. Hug, "HERMESS: Hybrid EMT/RMS Modern Electric power System Simulator," version 1.0.0, ETH Zurich Research Collection, 2026, doi: 10.3929/ethz-c-000805609.
and the paper describing the underlying models:
M. Katanic, J. Lygeros, and G. Hug, "Recursive dynamic state estimation for power systems with an incomplete nonlinear DAE model," IET Generation, Transmission & Distribution, vol. 18, no. 22, pp. 3657-3668, 2024, doi: 10.1049/gtd2.13308.
This software is free software, released by ETH Zurich under the GNU General Public License v3.0 or later (GPL-3.0-or-later). See LICENSE.txt for the full license text. It is distributed WITHOUT ANY WARRANTY; see the license for details.
hermess/tests/references/ additionally contains third-party benchmark data
redistributed under its own license (BSD-3-Clause, from the
PowerSimulationsDynamics.jl project); the license text and provenance live
next to the data.