Wind speed modeling for cascade clusters of wind turbines part 1: The cascade clusters of wind turbines. (15th August 2020)

• Record Type: Journal Article
• Authors: Dong, Xinghui; Li, Jia; Gao, Di; Zheng, Kai
• Is Part Of: Energy. Volume 205(2020)
• Publication Date: 2020-08-15
• Subjects: Cascade clusters of wind turbines -- Cascade characteristics -- Spectral clustering -- Segmentation algorithm; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2020.118097
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: Wind energy conversion efficiency has always been an important issue for wind farms. And wind speed calculation is the basic task and key work of wind energy conversion optimization. The cascade clusters of wind turbines are directly related to wind speed, and affected by the terrain, wake disturbance, location distribution and other factors. So it is very difficult to adopt parameter modeling. The cascade characteristics among cluster wind turbines (WTs) are embodied in historical operation data of the WTs. Taking the input wind direction as the initial parameter, we construct the WTs location correlation matrix of the neighborhood distribution relationship of WTs location; we then obtain the correlation relationship of the WTs production wind speed and power by combining the WTs production monitoring data. At the same time, "coupling element" and "aggregation element" WTs can be obtained from the cascade clusters. By verifying the data of a large wind farm, the model proposed in this paper clarifies the relationship between the wind speed and the cascade clusters; using this model, we can calculate the cluster distribution under different wind conditions. It is highly practical and can be applied to other wind farms to support formulation of the efficiency optimization strategies. Highlights: WT cascade correlation with distribution, wind speed and output power is studied. A segmentation algorithm is proposed which adopts the Spectral Clustering. CCWT is obtained, and factors that influence relationship of WTs are analyzed. CCWT under different wind direction are compared and CE and AE are introduced. Abstract.