The close coupling of artificial intelligence (AI) and electroencephalography (EEG) has substantially advanced human-computer interaction (HCI) technologies in the AI era. Different from traditional EEG systems, interpretability robustness AI-based systems are becoming particularly crucial. clarifies inner working mechanisms models thus can gain trust users. reflects AI's reliability against attacks perturbations, which is essential for sensitive fragile signals. Thus have attracted...

The dynamic behavior of the pump-turbine thrust bearing is important to safety operation unit. This paper analyzed lubrication and energy dissipation mechanism during load-rejection based on thermo-hydrodynamic model. results show that variation axial force maximum pad surface pressure basically consistent with inclination surface. change friction loss rotational speed, while temperature affected by combination speed. chaotic flow in oil tank accompanied different forms vortices, such as...

Face de-identification (De-ID) removes face identity information in images to avoid personal privacy leakage. Existing De-ID breaks the raw by cutting out regions and recovering corrupted via deep generators, which inevitably affect generation quality cannot control results according subsequent intelligent tasks (eg., facial expression recognition). In this work, for first attempt, we think from perspective of attribute editing propose an attribute-aware anonymization network (A3GAN)...

Compared to other modalities, EEG-based emotion recognition can intuitively respond the emotional patterns in human brain and, therefore, has become one of most concerning tasks brain-computer interfaces field. Since dependencies within regions are closely related emotion, a significant trend is develop Graph Neural Networks (GNNs) for recognition. However, region EEG have physiological bases that distinguish GNNs this field from those time series fields. Besides, there neither comprehensive...

Purpose The purpose of this paper is to analyze lubrication characteristics a tilting pad thrust bearing considering the effect thermal elastic deformation and collar. Design/methodology/approach This study used fluid–solid interaction (FSI) technique investigate for several typical operating conditions. influences rotational speed, load oil supply temperature on were analyzed. Findings three-dimensional (3D) film model clearly shows that there no pressure gradient but large gradients across...

The tribological performance of the piston ring–cylinder liner pair is affected by multiple factors. In this study, orthogonal test method used to comprehensively explore influence oil dilution, cylinder surface coating, and working conditions on friction wear pair. results show that a certain degree dilution beneficial performance, so further investigated using control variable method. depth decreases first then increases with increase dilution. Combined scanning electron microscope energy...

Purpose The purpose of this paper is to analyze lubrication characteristics a bidirectional thrust bearing in pumped storage, considering the effect thermal elastic deformation pad and collar. Design/methodology/approach This study used fluid–solid interaction (FSI) technique investigate for several typical operating conditions. influences conditions load on were analyzed. Then, various pivot eccentricities investigated effects position. Findings It found that radial tilt angle collar runner...

Purpose The purpose of this paper is to reveal the transient thermo-elasto-hydrodynamic lubrication mechanism a bidirectional thrust bearing in pumped-storage unit, and propose simulation method two-way fluid-solid-thermal interaction bearing. Design/methodology/approach used simulate three-dimensional bearing, during start-up shutdown process pumped storage unit. A pad including an oil hole modelled analyze temporal variation characteristics, such as film pressure, thickness temperature,...

The operating stability of guide bearings under load fluctuations in giant hydro turbines is a critical issue. As crucial component supporting rotating machinery and enduring radial forces, suffer from problems including rapid wear unstable operation. In this study, three-dimensional thermo-hydrodynamic model the 1000 MW Francis turbine unit bearing applied to analyze oil film lubrication characteristics flow tank. effects installation deviations on upper are analyzed, results demonstrate...

Sliding bearings are employed to support the rotor system and limit vibration amplitude. In high speed system, cavitation often occurs in oil film affects dynamic characteristics of sliding bearing greatly. this paper, numerical method is adopted simulate with homogeneous two-phase mixture flow using Singhal-et-al model commercial code FLUENT-solver. Cases without were also calculated at same time. Many computations different frequency ratios conducted. Then retrieved. The results show that...

Rain often poses inevitable threats to deep neural network (DNN) based perception systems, and a comprehensive investigation of the potential risks rain DNNs is great importance. However, it rather difficult collect or synthesize rainy images that can represent all situations would possibly occur in real world. To this end, paper, we start from new perspective propose combine two totally different studies, i.e., image synthesis adversarial attack. We first present an attack, with which could...

During turbocharger rotor speed-up tests, the sensor installed in compressor greatly affects dynamics. The effect of floating bearing stiffness coefficient on critical rotational speeds and shaft dynamic response due to unbalanced mass were both analyzed using dynamics theory a finite element analysis. results illustrate that initial design reduced 3rd order speed lower than maximum normal operating which leads rubbing failure during test. structure was then optimized reduce negative...

Drowsy driving has a crucial influence on safety, creating an urgent demand for driver drowsiness detection. Electroencephalogram (EEG) signal can accurately reflect the mental fatigue state and thus been widely studied in monitoring. However, raw EEG data is inherently noisy redundant, which neglected by existing works that just use single-channel or full-head channel model training, resulting limited performance of In this paper, we are first to propose Interpretability-guided Channel...

Deep image steganography is a data hiding technology that conceal in digital images via deep neural networks. However, existing methods only consider the visual similarity of container to host images, and neglect statistical security (stealthiness) images. Besides, they usually hides limited type thus relax constraint lossless extraction. In this paper, we address above issues unified manner, propose can embed with arbitrary types into for secure revealing. First, formulate as an...

The local optimality of motion vectors (MVs) is an intrinsic property in video coding, and any modifications to the MVs will inevitably destroy this optimality, making it a sensitive indicator steganography MV domain. Thus commonly used design steganalytic features, estimation for has become top priority steganalysis. However, existing works often estimated inaccurately or using unreasonable assumption, limiting its capability In paper, we propose estimate more reasonable comprehensive...

Drowsy driving has a crucial influence on safety, creating an urgent demand for driver drowsiness detection. Electroencephalogram (EEG) signal can accurately reflect the mental fatigue state and thus been widely studied in monitoring. However, raw EEG data is inherently noisy redundant, which neglected by existing works that just use single-channel or full-head channel model training, resulting limited performance of In this paper, we are first to propose Interpretability-guided Channel...

Abstract Thrust bearing is a key component of hydroelectric generator and bears the axial load unit. During transient processes, such as start stop unit, thrust usually occurs wear phenomenon at pad surface, which affects stable operation unit reduces service life bearing. Based on two-way fluid-solid-thermal interaction, this paper did thermal-elastic-hydrodynamic (TEHD) analysis tilting under pump mode. For fluid region, Reynolds equation viscosity-temperature are solved. solid heat...

The dynamic characteristics of the shaft system for a dredging pump were studied with Finite Element Method (FEM) by SAMCEF ROTOR. At first, influence fluid-solid coupling interaction mud water and impeller, sealing on lateral critical speeds analyzed. results indicated that must be taken into consideration, while need not to. Then effects radial thrust rolling bearings discussed, which shows bearing close to impeller has greatest impact 1st order speed. last, upper lower limits lateral,...