MATH_code : False Data Injection Attack Detection in Smart Grids based on Reservoir Computing

<b>1_matlab_data_generation/</b><br>MATLAB scripts used to generate the synthetic datasets based on the IEEE 57-bus system. These scripts simulate clean power flow measurements and apply targeted FDIA manipulations to create labeled datasets for training and evaluation.<b>2_familiarize_with_models.ipynb</b><br>A preliminary notebook used to explore and understand the behavior and structure of various machine learning models in the context of FDIA detection.<b>3_literature_analysis_and_mapping.ipynb</b><br>Contains the Python code used for executing the systematic mapping study (SMS), including automated processing of literature data and thematic clustering.<b>4_final_models_pipeline.ipynb</b><br>The final implementation pipeline that loads the data, applies preprocessing and encoding (e.g., latency or ISI), trains the detection models, and stores performance metrics.<b>5_statistical_analysis.ipynb</b><br>Jupyter notebook that conducts the full statistical evaluation of all trained models, including hypothesis testing (Shapiro-Wilk, Friedman, Wilcoxon) and visualization of key performance metrics.<b>datasets_and_case/</b><br>Directory containing all 21 generated datasets (plus the base case). These include manipulated datasets with varying attack intensities and compromised meter counts.<b>literature_data/</b><br>CSV and text exports of the literature search results from scientific databases, used as the basis for the systematic mapping study.<br>rc_all_models_final_results.csv