In recent years, wind energy has increased its participation in the world energy mix. Besides its advantages, wind energy is not constant and presents undesired fluctuations, which can affect the power quality, reliability, and generation dispatch. Energy storage systems (ESS) are used to smooth the wind power output, reducing fluctuations. Within the variety of energy storage systems available, the battery energy storage system (BESS) is the most utilized to smooth wind power output. However, the capacity of BESS to compensate for fluctuations is usually exceptionally large, which will increase the capital cost of the system and reducing its suitability. To solve this problem, some studies focused on implementing control systems to optimize BESS and reduce its required size. This paper presents a literature review of the control strategies that use the battery energy storage systems to smooth the wind power output, which can guide future practical applications. Based on this review we found that most of the studies use PI, Fuzzy, and MPC control strategies but no many studies focusing on deep learning which is an arising control technology in the field of Wind Energy.

近年来，风能已越来越多地参与世界能源组合。除了其优点之外，风能不是恒定的，并且会出现不希望的波动，这会影响电能质量，可靠性和发电调度。储能系统（ESS）用于平滑风能输出，减少波动。在各种可用的储能系统中，电池储能系统（BESS）最能使风能输出平稳。但是，BESS补偿波动的能力通常非常大，这将增加系统的资本成本并降低其适用性。为了解决这个问题，一些研究集中于实施控制系统以优化BESS并减小其所需的尺寸。本文介绍了使用电池能量存储系统来平滑风能输出的控制策略的文献综述，可为将来的实际应用提供指导。根据此评论，我们发现大多数研究都使用PI，Fuzzy和MPC控制策略，但很少有研究专注于深度学习，而深度学习是风能领域中一种新兴的控制技术。