Blade attachment degree classification for marine current turbines using AGMF-DFA under instantaneous variable current speed

Abstract

Marine current turbine (MCT) blades are continuously exposed to seawater, making them extremely susceptible to marine biological attachment and causing MCT blade imbalance fault. It is challenging to extract attachment features with the instantaneous variable current speed. In fact, the collected signal is highly nonlinear and nonstationary, which involves different scales and dimensions affected by the instantaneous current speed. In this paper, improved multifractal detrended fluctuation analysis (MF-DFA) is adopted to analyze such signals. The original MF-DFA is easily affected by the order of the fitting polynomial and cannot achieve the desired adaptive effect. The paper combines grey relation analysis and ensemble empirical mode decomposition (EEMD) to adaptively select the detrended function. The proposed adaptive grey MF-DFA (AGMF-DFA) method is validated with data collected from a testbed composed of an MCT coupled to a 230W permanent magnet synchronous generator. The result shows that the proposed method is effective for the attachment degree classification under the instantaneous variable current speed.