基于深度学习的风力发电机组故障预警方法研究综述

风能作为重要的可再生能源，近几十年来，全球风能使用规模迅速增长，陆上和海上风力发电机组发电容量不断增加。由于风力发电机组故障维修成本巨大，因此必须开发有效且可靠的风力发电机组故障预警方法，在风电机组发生故障前进行提前预警，以便降低风电场的运营和维护成本。目前风电机组数据采集与监视控制系统(supervisory control and data acquisition, SCADA)已经在风电场有了广泛的应用，其中蕴含着大量的潜在数据信息，同时深度学习方法在海量数据挖掘方面有比较明显的优势，因此深度学习方法在风力发电机组故障预警领域的应用潜力巨大。综述了近年来相关深度学习方法在风力发电机组故障预警的研究进展，总结了风电机组故障预警的大体步骤，分析了各个步骤的具体处理方法，对每种技术方法的特点进行整理分析。最后阐述了深度学习在风电机组故障预警领域所面临的挑战，并对今后的研究重点进行了展望。

A review of Wind Turbine fault warning methods Based on deep learning

Wind energy is vital renewable energy, in recent decades, global wind energy use has scaled rapidly, and onshore and offshore wind turbine generating capacity is increasing. Because the wind turbine maintenance cost is enormous, so we must develop effective and reliable wind turbine early warning methods to early warning before wind turbine failure, in order to reduce wind farm operation and maintenance costs. Supervisory Control And Data Acquisition (SCADA) has been widely used in wind farms, containing much potential Data information. At the same time, the deep learning method has apparent advantages in mass data mining, so the deep learning method has great potential in the field of wind turbine fault warning. This paper reviews the research progress of deep learning methods in wind turbine fault warning in recent years, summarizes the general steps of wind turbine fault warning, analyzes the specific processing methods of each step, and sorts out and analyzes the characteristics of each method. Finally, the challenges of deep learning in wind turbine fault warning are described, and future research priorities are forecasted.