Detecting rotating machinery faults under different working conditions with cross-domain negative correlated ensemble algorithm. (November 2021)

• Record Type: Journal Article
• Authors: Pang, Shan; Wang, Jinglin; Yang, Xinyi; Zhang, Xiaofeng
• Is Part Of: Measurement. Volume 184(2021)
• Publication Date: 2021-11
• Subjects: Ensemble method -- Extreme learning machine -- Deep learning -- Fault diagnosis; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2021.109951
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: A novel ensemble algorithm is proposed to solve the problem of diagnosis under different conditions. DCELM with fast training speed is used as base learner to extract features from time-frequency images. The algorithm can select base learners which are negatively correlated in terms of their outputs. By exploiting unlabeled target domain data, our ensemble achieves "cross-condition" capability. The effectiveness of the algorithm is evaluated by experiments on gears, rotor and rolling bearing. Abstract: When detecting faults of rotating machinery, variation in working conditions leads to the distribution mismatch between training and test data. Thus, the diagnosis accuracy of existing deep learning models deteriorates greatly. To address that, we propose a novel ensemble algorithm. The ensemble algorithm employs deep convolutional extreme learning machine (DCELM) as base learners to learn representative features from time-frequency images. Then, a selection scheme is designed to select diverse base learners which are negatively correlated. Meanwhile, by calculating the correlation term using both the source and the target domain data, the attained ensemble diversity is also valid in target domains. The proposed ensemble algorithm is applied to detect faults of rotating machinery components in three cases. Results show that it attains the best performance on all cross-condition tasks. On average, its accuracy is 1.9% higher than other state-of-the-art ensemble methods and achieves 3.6% enhancement compared with its base learner.