文章摘要
周念成,廖建权,王强钢,等.深度学习在智能电网中的应用现状分析与展望[J].电力系统自动化. DOI: 10.7500/AEPS20180323002.
ZHOU Niancheng,LIAO Jianquan,WANG Qianggang, et al.Developing Status and Prospect Analysis of Deep Learning in Smart Grid[J].Automation of Electric Power Systems. DOI: 10.7500/AEPS20180323002.
深度学习在智能电网中的应用现状分析与展望
Developing Status and Prospect Analysis of Deep Learning in Smart Grid
DOI:10.7500/AEPS20180323002
关键词: 人工智能  大数据  深度学习  智能电网  可再生能源  电力信息物理系统
KeyWords: Artificial Intelligence  Big Data  Deep Learning  Smart Grid  Renewable Energy  Cyber-physical Power System
上网日期:2018-11-30
基金项目:国家自然科学基金项目(面上项目,重点项目,重大项目)
作者单位E-mail
周念成 输配电装备及系统安全与新技术国家重点实验室重庆大学 cee_nczhou@cqu.edu.cn 
廖建权 输配电装备及系统安全与新技术国家重点实验室重庆大学 jquanliao@cqu.edu.cn 
王强钢 输配电装备及系统安全与新技术国家重点实验室重庆大学 qianggang1987@cqu.edu.cn 
李春艳 输配电装备及系统安全与新技术国家重点实验室重庆大学 lichunyan59@163.com 
李剑 输配电装备及系统安全与新技术国家重点实验室重庆大学 lijian@cqu.edu.cn 
摘要:
      深度学习是机器学习研究中的一个新领域,其强大的数据分析、预测、分类能力契合智能电网中大数据应用的需求。本文首先总结了深度学习基本思想,介绍深度学习的5种模型(生成式对抗网络、递归神经网络、卷积神经网络、堆叠自动编码器和深度信念网络)的结构、基本原理、训练方法,概括其应用特征。综述了电力系统中的故障诊断、暂态稳定性分析、负荷及新能源功率预测、运行调控等应用深度学习技术的研究现状。针对深度学习的技术特点,结合电力系统各生产环节,构建深度学习技术在电力系统中的应用框架。最后,从多能源系统运行调控、电力电子化系统安全分析、柔性设备故障诊断、电力信息物理系统的安全防护等方面对深度学习应用进行展望。
Abstract:
      Deep learning is a new field of machine learning. Its powerful data analysis, prediction, and classification capabilities fit the needs of big data applications in smart grid. This paper summarized the basic ideas of deep learning, introduced the structures, basic principles, and training methods of five models of deep learning (Generative Adversarial Networks, Recurrent Neural Network, Convolution Neural Network, Stacked Auto Encoder, Deep Belief Network), and summarized their application characteristics. The applications of deep learning techniques in fault diagnosis, transient stability analysis, load forecasting, control and optimization in power system are summarized. Based on the technical characteristics of deep learning, combined with the production processes of the power system, the application framework for deep learning technology in the power system was constructed. Finally, the application of deep learning was prospected from the aspects of multi-energy system operation regulation, power electronic system security analysis, flexible equipment fault diagnosis, and Cyber-physical power system security protection.
查看全文   查看/发表评论  下载PDF阅读器