Power Curve Modeling for Wind Turbine Using Hybrid-driven Outlier Detection Method

• DOI: 10.35833/mpce.2021.000769
• Publication Date: 2023-07-26T06:21:40Z
• Authors: Qi Yao; Yang Hu; Jizhen Liu; Tianyang Zhao; Xiao Qi; and Shanxun Sun
• Keywords: Data pre-processing; Curve fitting

Wind power curve modeling is essential in the analysis and control of wind turbines (WTs), data preprocessing a critical step accurate modeling. As traditional methods do not sufficiently consider WT models, this paper proposes new cleaning method for In method, model-data hybrid-driven (MDHD) outlier detection algorithm constructed, an adaptive update rule major parameters designed based on model mechanism. Simultaneously, because MDHD considers multiple types operating WTs, anomaly results require further analysis. Accordingly, expert system developed which knowledgebase inference engine are coupling relationships different data. Finally, abnormal eliminated completed. The proposed compared numerical cases, superiority demonstrated.