Machine Learning Package for ContGridMod

Implementation of physics-informed machine learning routines for the continuous power grid model package ContGridMod.

Installation

ContGridModML can easily be installed using the Julia package manager.

julia> using Pkg
julia> Pkg.add(url="https://github.com/laurentpagnier/ContGridMod.jl#FiniteElements")
julia> Pkg.add(url="https://github.com/julianfritzsch/ContGridModML.jl")

Note

This package needs the FiniteElements branch of ContGridMod. This might conflict with preinstalled versions of ContGridMod. If there are any problems using this package, remove any versions of ContGridMod checked out for development and try again.

Quickstart Guide

To reproduce the results you only need to run two functions. The static parameters, i.e., the susceptances in $x$ and $y$ direction ($b_x(\mathbf{r})$ and $b_y(\mathbf{r})$), can be learned by running

julia> sol = learn_susceptances()

Similarly, to learn the dynamical parameters, i.e., the inertia $m(\mathbf{r})$ and the damping $d(\mathbf{r})$, run

julia> sol = learn_dynamical_parameters()