FELIPE WHITAKER DE A. M. TAVARES
Título: Short Term Wind Speed Scenario Generation for Brazil with Improved Generative Adversarial Networks
Data: 18/09/2024, 15:00h
Sala Zoom : https://puc-rio.zoom.us/j/98631434540?pwd=HMeu8uk5kHhNC2PB2iSB99tE3UV94j.1
Orientadores: Fernando Luiz Cyrino Oliveira | PUC-Rio & Marley Vellasco | PUC-Rio
Resumo: The variability of renewable energy sources, such as wind power, presents a significant challenge for grid operators in maintaining operational stability. This is specially true to the medium-term (from hours to days ahead), which is both influenced by recent past data and broader trends, and heavily influences decision making. This research proposes a fully Convolutional Generator Network conditioned on the previous step of u- and v- wind speed components to generate wind speed scenarios using the Conditional Generative Adversarial Networks training algorithm. The model is compared to the state of the art in weather forecasting, Numerical Weather Prediction Systems. The model is shown to outperform the benchmark with much lower computational cost, less input data and similar long-term stability; achieving better results for a third of the comparisons in the test dataset when predicting an entire month (56 12-hourly steps) from a single data point.