An entropy clustering method for blades biofouling detection of marine current turbine under variable marine current speeds

Marine current turbine (MCT) attracts extensive attention as a renewable energy power generation equipment. However, accurate and reliable blades biofouling detection for MCT remains a significant challenge, especially under variable marine current speed. This paper proposes an entropy clustering method to improve the detection accuracy of blades biofouling for MCT under variable marine current speed. Firstly, the stator current signal is decomposed based on variational mode decomposition (VMD) to enlarge the feature of blades biofouling. Secondly, to eliminate the influence of swell effect, fluctuating features caused by swell effect are divided into each subspace based on the proposed entropy clustering method. At the last stage, principal component analysis models (PCA) are established in each feature subspace to detect blades biofouling. To verify the effectiveness of the proposed method, experiments are carried out on a 230W MCT platform. The results show that the proposed method has satisfactory performance for blades biofouling detection under variable marine current speed.