Despite the recent advances on intelligent fault diagnosis of rolling element bearings, existing research works mostly assume training and testing data are drawn from same distribution. However, due to variation operating condition, domain shift phenomenon generally exists, which results in significant performance deterioration. To address cross-domain problems, latest preferably apply adaptation techniques marginal distributions. it is usually assumed that sufficient available for training,...

This work demonstrates the controllable synthesis of cobalt sulfide hollow nanospheres and phase-dependent catalytic properties for OER HER.

Rotating machinery fault diagnosis problems have been well-addressed when sufficient supervised data of the tested machine are available using latest data-driven methods. However, it is still challenging to develop effective diagnostic method with insufficient training data, which highly demanded in real-industrial scenarios, since high-quality usually difficult and expensive collect. Considering underlying similarities rotating machines, mining on different but related equipments...

In the past years, practical cross-domain machinery fault diagnosis problems have been attracting growing attention, where training and testing data are collected from different operating conditions. The recent advances in closed-set domain adaptation well addressed basic problem mode sets identical source target domains. While some attempts also made on partial open-set adaptations, no prior information of target-domain modes can be usually available real industries, that forms a...

Data-driven machinery fault diagnosis methods have been successfully developed in the past decades. However, cross-domain diagnostic problems not well addressed, where training and testing data are collected under different operating conditions. Recently, domain adaptation approaches popularly used to bridge this gap, which extract domain-invariant features for diagnostics. Despite effectiveness, most existing assume label spaces of identical that indicates mode sets same scenarios. In...

In the recent years, data-driven machinery fault diagnostic methods have been successfully developed, and tasks where training testing data are from same distribution well addressed. However, due to sensor malfunctions, can be collected at different places of machines, resulting in feature space with significant discrepancy. This challenging issue has received less attention current literature, existing approaches generally fail such scenarios. article proposes a domain adaptation method for...

DC microgrids consist of multiple power electronic converter units interconnected in a network with sources and loads. They are commonly found the systems electric ships, aircrafts, etc. The main focus this paper is to study stability converters-based dc microgrids. These high switching frequency electronics controlled way as maintain constant voltage, current, or load. Due their bandwidth, they can be simplified For reason, nonlinear system arises where it no longer satisfactory assume...

In the past years, deep learning-based machinery fault diagnosis methods have been successfully developed, and basic diagnostic problems well addressed where training testing data are collected under same operating conditions. When from different distributions, domain adaptation approaches introduced. However, existing generally assume availability of target-domain in all health conditions during training, which is not accordance with real industrial scenarios. This article proposes a method...

Intelligent data-driven machinery fault diagnosis methods have been popularly developed in the past years. While fairly high accuracies obtained, large amounts of labeled training data are mostly required, which difficult to collect practice. The promising collaborative model solution with multiple users poses demands on privacy due conflict interests. Furthermore, real industries, from different can be usually collected machine operating conditions. domain shift phenomenon and concern make...

Mainstream lane marker detection methods are implemented by predicting the overall structure and deriving parametric curves through post-processing. Complex line shapes require high-dimensional output of CNNs to model global structures, which further increases demand for capacity training data. In contrast, locality a has finite geometric variations spatial coverage. We propose novel solution, FOLOLane, that focuses on modeling local patterns achieving prediction structures in bottom-up...