基于人工智能算法的电力系统负荷预测研究综述
A Review of Power System Load Forecasting Based on Artificial Intelligence Algorithms
杨雷 1罗雪红 1韩鹍 1张启立 2郭鹏 2李晓飞2
作者信息
- 1. 国网渭南供电公司,陕西 渭南 714000
- 2. 北京国电通网络技术有限公司,北京 100070
- 折叠
摘要
在能源互联网大环境下,风光等新能源发电的大量并网带来的间歇性问题影响了电力系统的稳定运行.传统的电力负荷预测方法对于该情况下的动态负荷精度已无法保证,而基于人工智能算法的预测方法得到了广泛的应用.对此,首先介绍了电力系统负荷预测方法的分类和必要性,并将基于人工智能算法的电力系统负荷预测分为基于传统机器学习、基于深度学习、基于组合模型3种方法展开综述,最后对电力负荷预测领域的未来发展进行了展望和总结.
Abstract
Under the environment of energy internet,the intermittency problem brought by large-scale integration of new energy sources such as wind and photovoltaic generations impacts stable operation of the power system.The conventional load forecasting methods can no longer guarantee their accuracy for this situation,and the forecasting methods based on artificial intelligence algorithms have found wide application.This paper encompasses first an outline on classification and necessity of power system load forecasting methods,and then a summary of the AI algorithm-based methods from the perspectives of different algorithm classes,i.e.,conventional machine learning,deep learning,and combined models.The paper ends with a future outlook of this fast advancing field.
关键词
电力系统负荷预测/传统机器学习/深度学习/组合模型Key words
power system load forecasting/conventional machine learning/deep learning/combinatorial model引用本文复制引用
出版年
2024
NETL
NSTL
万方数据