An Open Source Software Stack for Tuning the Dynamical Behavior of Complex Power Systems

TL;DR

This paper introduces BlockSystems.jl and NetworkDynamics.jl, open-source Julia packages for efficient, detailed transient stability simulations of power networks, enabling easy modeling, control tuning, and integration of machine learning techniques.

Contribution

The paper presents novel Julia software packages that simplify power system modeling and control tuning, incorporating machine learning and automatic differentiation capabilities.

Findings

Successful implementation of the Nordic5 test case.

Efficient tuning of control parameters using machine learning.

Open-source ecosystem with state-of-the-art solvers.

Abstract

BlockSystems.jl and NetworkDynamics.jl are two novel software packages which facilitate highly efficient transient stability simulations of power networks. Users may specify inputs and power system design in a convenient modular and equation-based manner without compromising on speed or model detail. Written in the high-level, high-performance programming language Julia a rich open-source package ecosystem is available, which provides state-of-the-art solvers and machine learning algorithms. Motivated by the recent interest in the Nordic inertia challenge we have implemented the Nordic5 test case and tuned its control parameters by making use of the machine learning and automatic differentiation capabilities of our software stack.