Energy demand forecasting using convolutional neural network and modified war strategy optimization algorithm.

Area of Science:

• Energy Systems Engineering
• Artificial Intelligence
• Data Science

Background:

• Effective electricity demand prediction is vital for energy industry planning and government policy.
• Traditional methods struggle with complex patterns, necessitating advanced analytical tools.
• Machine learning offers a powerful alternative for enhancing energy demand forecasting accuracy.

Purpose of the Study:

• To introduce a novel hybrid model, the Modified War Strategy Optimization-Based Convolutional Neural Network (MWSO-CNN), for precise electricity demand prediction.
• To leverage the strengths of convolutional neural networks (CNNs) and a modified war strategy optimization (MWSO) technique for improved forecasting.
• To optimize CNN hyperparameters using MWSO for enhanced predictive performance.

Main Methods:

• Development of the MWSO-CNN model integrating MWSO for CNN hyperparameter tuning.
• Application of the MWSO-CNN model to a real-world electricity demand dataset.
• Comparative analysis against state-of-the-art machine learning techniques.

Main Results:

• The MWSO-CNN model demonstrated superior performance in electricity demand prediction compared to existing methods.
• Optimized CNN hyperparameters through MWSO significantly enhanced prediction precision.
• Validation on a real-world dataset confirmed the model's effectiveness and robustness.

Conclusions:

• The MWSO-CNN approach provides a highly accurate and efficient solution for electricity demand forecasting.
• This method offers a cost-effective strategy for energy consumption prediction, benefiting the energy sector and society.
• The study highlights the potential of hybrid optimization and deep learning techniques in energy management.