Bibliographic citations
Sánchez, N., (2023). Desarrollo de un modelo de pronóstico de energía a corto plazo, basado en redes neuronales recurrentes, para mejorar la programación de generación de energía en centrales eólicas [Trabajo de suficiencia profesional, Universidad Nacional de Ingeniería]. http://hdl.handle.net/20.500.14076/27051
Sánchez, N., Desarrollo de un modelo de pronóstico de energía a corto plazo, basado en redes neuronales recurrentes, para mejorar la programación de generación de energía en centrales eólicas [Trabajo de suficiencia profesional]. PE: Universidad Nacional de Ingeniería; 2023. http://hdl.handle.net/20.500.14076/27051
@misc{renati/712831,
title = "Desarrollo de un modelo de pronóstico de energía a corto plazo, basado en redes neuronales recurrentes, para mejorar la programación de generación de energía en centrales eólicas",
author = "Sánchez Huayana, Nils Ericsson",
publisher = "Universidad Nacional de Ingeniería",
year = "2023"
}
The aim of this research study is to enhance the accuracy of short-term wind energy forecasting by proposing a forecasting model based on Variational Mode Decomposition and Long Short-Term Memory (VMD-LSTM). The data used for this model were sourced from a wind farm located in Talara, in the north of Peru. Initially, the Variational Mode Decomposition method was adopted to decompose the wind energy data into three constituent modes: the long-term component, the fluctuation component, and the random component. Subsequently, the Recurrent Neural Network (RNN), specifically the Long Short-Term Memory (LSTM) variant, was employed to deeply analyze the characteristics of these three constituent modes. Thanks to its unique forget gate and memory gate structure, the LSTM learns the association with long-term time series to construct a multi- step forecasting model. The algorithm was developed using the Python programming language in the Google Colab development environment, aiming to leverage its computational power and user-friendly interface. The accuracy of the VMD-LSTM model was evaluated, and the experimental results indicate that the proposed model provides superior performance in multi-step forecasting.
This item is licensed under a Creative Commons License