电工技术学报2017，Vol.32Issue(22)：233-241,9.DOI:10.19595/j.cnki.1000-6753.tces.160694
基于自适应EEMD和分层分形维数的风电机组行星齿轮箱故障检测
Diagnosis of Wind Turbine Planetary Gearbox Faults Based on Adaptive EEMD and Hierarchical Fractal Dimension
摘要
Abstract
The main defect of the traditional ensemble empirical mode decomposition (EEMD) is that the important parameters of the added white noise are set by artificial experience. In the paper, an adaptive EEMD is proposed based on the study of the factors that caused the modal aliasing. This method could set the parameters for different signals adaptively to achieve optimal decomposition effectiveness. Firstly, the singular value decomposition (SVD) was used to decompose and reconstruct the signals. Next, the reconstruction signals were used to determine the parameters of the white noise adaptively. Finally, using the proposed method, the signals were decomposed to a series of intrinsic mode function (IMF). Fractal dimension is good for the evaluation of the characteristics of IMF, so it is effective to identify different types of vibration signals. The hierarchical fractal dimension was used to improve the accuracy and efficiency of signal recognition. The experimental and simulation results of the gearbox of the wind turbine show that the proposed method is more effective compared with the existing techniques.关键词
风电机组/行星齿轮箱故障诊断/自适应平均经验模态分解/分层分形维数Key words
Wind turbines/diagnosis of planetary gearbox faults/adaptive ensemble empirical mode decomposition (EEMD)/hierarchical fractal分类
信息技术与安全科学引用本文复制引用
李东东,周文磊,郑小霞,王浩..基于自适应EEMD和分层分形维数的风电机组行星齿轮箱故障检测[J].电工技术学报,2017,32(22):233-241,9.基金项目
国家自然科学基金(51507098,51507100)、上海市人才发展基金(201365)和上海市科委资助项目(15YF1404600,13DZ2251900, 10DZ2273400). （51507098,51507100）
