Due to frequent changes in wind speed and wind direction, the yaw system needs to be frequently aligned with the wind direction, which may cause a series of engineering problems such as downtime caused by frequent actions. In this paper, SCADA data is used to filter out abnormal data such as shutdown and power limit through empirical analysis. Then the DBSCAN algorithm is used to filter the data to obtain a better SCADA data set. The data set is modeled by curved surface polynomial least squares. Through the analysis of the model, it is concluded that the yaw coefficient and wind speed have different influence characteristics on power. The yaw coefficient above rated wind speed has less influence on output power, and the yaw of different wind speed has different influence on power.

Date Published: 13 November 2019
PDF: 8 pages
Proc. SPIE 11343, Ninth International Symposium on Precision Mechanical Measurements, 1134319 (13 November 2019); doi: 10.1117/12.2548698

Published in SPIE Proceedings Vol. 11343:
Ninth International Symposium on Precision Mechanical Measurements
Liandong Yu, Editor(s)