<![CDATA[This study aims to improve the accuracy of energy consumption prediction using the Bidirectional Long Short-Term Memory (BLSTM) model which is known to be able to handle temporal dependencies in time series data. However, the performance of BLSTM is greatly affected by the hyperparameter configuration, which often requires manual tuning which is inefficient. To address this, this study proposes an optimization framework that combines BLSTM with Chaotic Particle Swarm Optimization (CPSO) to automatically adjust hyperparameters such as the number of hidden units and learning rate. Experiments show that BLSTM optimized with CPSO produces higher prediction accuracy compared to traditional methods such as grid search and random search. By utilizing the chaos map, CPSO improves exploration and exploitation capabilities, accelerates convergence, and finds more optimal solutions. The integration of CPSO and BLSTM shows promising results for improving the performance of time series prediction models, especially in energy consumption forecasting.]]>
Copyrights © 2024
