Supplementary data: "Revealing interactions between HVDC cross-area flows and frequency stability with explainable AI"

This repository contains processed data and result files for the paper Revealing interactions between HVDC cross-area flows and frequency stability with explainable AI. The code for producing the processed data and the results is available at github.

Data

The data folder contains the feature and target data used to train the ML model:

stability_input: A Folder containing training and test sets for the stability model for each area.

flow_input: A Folder containing training and test sets for the flow model for each link.

raw_input_data.h5 : The aggregated external features without additional engineered features.

input_forecast.h5 and input_actual.h5: The day-ahead available (forecast) and ex-post available (actual) data of external features including the engineered features.

indicators.h5 :  The grid frequency stability indicators.

documentation_of_data_download: Information files concerning the ENTSO-E raw data and its aggregation.

HVDClinks: A Folder containing preprocessed time series for scheduled and unscheduled HVDC flows

Data sources

Most of the data is derived from ENTSO-E Transparency Platform data [1]. The grid stability indicators (indicators.h5) are based on publicly available data from the German Transmission System Operators (TSOs) [2].

Results

The stability_results and the flow_results folder contain the results of hyperparameter optimization, model prediction and interpretation via SHAP for the respective models.

cv_results_gtb_full.csv : Performance results for each combination in the hyperparameter optimization.

cv_best_params_gtb_full.csv : Hyperparameters used in the final (optimized) model.

shap_values_gtb_full.npy : First-order SHAP values calculated on different data sets: The train set, the randomized test set and the continuous test set.

y_pred.h5/y_pred_links.h5 : Predictions of daily profile predictor and Machine Learning models.

Disclaimer

The data might be subject to copyright or related rights. Please consult the primary data owner.