Wind Turbine Abnormal Data Identification Based on MKIF Model

The identification of wind turbine abnormal data is the basis for subsequent wind power prediction and health assessment of wind turbine. This paper analyzes the characteristics of wind turbine abnormal data in detail, and designs the MKIF (Mini batch K Means-Isolation Forest) algorithm to realize the identification of abnormal data of wind turbine. The MKIF algorithm introduces Mini batch K Means clustering into the partition process of the Isolation Forest search tree, and use Silhouette Coefficient to supervise the number and location of the split nodes of the tree. The MKIF anomaly score is defined to describe the degree of isolation of the data, which can effectively identify and eliminate abnormal data and use normal data to establish the main power band. Taking the actual operating data of two wind turbines as an example, the effectiveness of the MKIF abnormal data identification algorithm is verified.