Hu Huanhuan, Gong Shufen, Taheri Bahman
College of Big Data and Artificial Intelligence, Chizhou University, Chizhou, 247100, Anhui, China.
Science and Research Branch, Islamic Azad University, Tehran, Iran.
Heliyon. 2024 Feb 29;10(6):e27353. doi: 10.1016/j.heliyon.2024.e27353. eCollection 2024 Mar 30.
Predicting the electricity demand is a key responsibility for the energy industry and governments in order to provide an effective and dependable energy supply. Traditional projection techniques frequently rely on mathematical models, which are limited in their ability to recognize complex patterns and correlations in data. Machine learning has emerged as a viable tool for estimating electricity in the last decade. In this study, the Modified War Strategy Optimization-Based Convolutional Neural Network (MWSO-CNN) has been provided for electricity demand prediction. To increase the precision of electricity demand prediction, the MWSO-CNN approach integrates the benefits of the modified war strategy optimization technique and the convolutional neural network. To improve efficiency, the modified war strategy optimization technique is employed to adjust the hyperparameters of the CNN algorithm. The suggested MWSO-CNN approach is tested on a real-world electricity demand dataset, and the findings show that it outperforms many state-of-the-art machine learning techniques for predicting electricity demand. In general, the suggested MWSO-CNN approach could offer a successful and cost-effective strategy for predicting energy consumption, which will benefit both the energy business and society as a whole.
预测电力需求是能源行业和政府的一项关键职责，以便提供有效且可靠的能源供应。传统的预测技术通常依赖数学模型，而这些模型在识别数据中的复杂模式和相关性方面能力有限。在过去十年中，机器学习已成为一种可行的电力预测工具。在本研究中，提出了基于改进战争策略优化的卷积神经网络（MWSO-CNN）用于电力需求预测。为了提高电力需求预测的精度，MWSO-CNN方法融合了改进战争策略优化技术和卷积神经网络的优点。为了提高效率，采用改进战争策略优化技术来调整CNN算法的超参数。所提出的MWSO-CNN方法在真实世界的电力需求数据集上进行了测试，结果表明它在预测电力需求方面优于许多先进的机器学习技术。总体而言，所提出的MWSO-CNN方法可为预测能源消耗提供一种成功且具有成本效益的策略，这将使能源行业和整个社会都受益。
