Siyuan Liu^1,*, Jinying Huang², Jiancheng Ma¹, Licheng Jing², Yuxuan Wang²

CMC-Computers, Materials & Continua, Vol.79, No.1, pp. 761-777, 2024, DOI:10.32604/cmc.2024.049484

Abstract Recent research in cross-domain intelligence fault diagnosis of machinery still has some problems, such as relatively ideal speed conditions and sample conditions. In engineering practice, the rotational speed of the machine is often transient and time-varying, which makes the sample annotation increasingly expensive. Meanwhile, the number of samples collected from different health states is often unbalanced. To deal with the above challenges, a complementary-label (CL) adversarial domain adaptation fault diagnosis network (CLADAN) is proposed under time-varying rotational speed and weakly-supervised conditions. In the weakly supervised learning condition, machine prior information is used for sample annotation via cost-friendly complementary label learning.… More >

L. Cheng^a , Y. L. Zhang^a,† , J. F. Li^b

Frontiers in Heat and Mass Transfer, Vol.19, pp. 1-7, 2022, DOI:10.5098/hmt.19.32

Abstract The internal characteristics during rotational speed fluctuation have an important influence on centrifugal pump to avoid or utilize its transient performance. In this paper, a circulation piping system that includes a low-specific-speed centrifugal pump is established to study the energy distribution characteristics in a centrifugal pump during speed fluctuation. The unsteady flow in the entire system is numerically calculated with a user-defined function, the sliding grid method, and the RNG k-ε turbulence model. Then, the energy distribution of the transient flow field in the centrifugal pump model during speed fluctuation is diagnosed with vortex dynamics by using flow section and… More >