Long Short‐Term Memory Wavelet Neural Network for Renewable Energy Generation Forecasting

• DOI: 10.1155/int/8890906
• Publication Date: 2025-01-31T11:10:38Z
• Authors: Eliana Vivas; Héctor Allende-Cid; Lelys Bravo de Gu...; Aurelio F. Bariviera; Rodrigo Salas

Renewable energy forecasting is crucial for pollution prevention, management, and long‐term sustainability. In response to the challenges associated with forecasting, simultaneous deployment of several data‐processing approaches has been used in a variety studies order improve energy–time‐series analysis, finding that, when combined wavelet deep learning techniques can achieve high accuracy applications. Consequently, we investigate implementation various wavelets within structure long short‐term memory neural network (LSTM), resulting new LSTM (LSTMW) network. addition, as an improvement phase, modeled uncertainty incorporated it into forecast so that systemic biases deviations could be accounted (LSTMW luster: LSTMWL). The models were evaluated using data from six renewable power generation plants Chile. When compared other approaches, experimental results show our method provides prediction error acceptable range, achieving coefficient determination ( R 2 ) between 0.73 0.98 across different test scenarios, consistent alignment forecasted observed values, particularly during first 3 steps.