VMD-LSTM Wind Power Forecasting: Hybrid Framework for Northeast Brazil

Hybrid VMD-LSTM framework for short-term wind power forecasting in Northeast Brazil. This repository contains the code for the paper: "Hybrid VMD-LSTM Model for Short-Term Wind Power Forecasting: A Case Study of Northeast Brazil" submitted to Applied Energy. Key features:- Curtailment data reconstruction methodology- Variational Mode Decomposition (VMD) with K=6 modes- LSTM architecture with dropout and L2 regularization- Multi-horizon forecasting (1h, 6h, 12h, 24h) Results: MAPE = 3.16%, R² = 0.9879 for 6-hour horizon Data source: Brazilian National System Operator (ONS) - https://dados.ons.org.br/