• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.4
• /
• pp.1275-1292
• /
• 2021

As a fine-grained classification problem, aspect-level sentiment classification predicts the sentiment polarity for different aspects in context. To address this issue, researchers have widely used attention mechanisms to abstract the relationship between context and aspects. Still, it is difficult to effectively obtain a more profound semantic representation, and the strong correlation between local context features and the aspect-based sentiment is rarely considered. In this paper, a hybrid attention capsule network for aspect-level sentiment classification (ABASCap) was proposed. In this model, the multi-head self-attention was improved, and a context mask mechanism based on adjustable context window was proposed, so as to effectively obtain the internal association between aspects and context. Moreover, the dynamic routing algorithm and activation function in capsule network were optimized to meet the task requirements. Finally, sufficient experiments were conducted on three benchmark datasets in different domains. Compared with other baseline models, ABASCap achieved better classification results, and outperformed the state-of-the-art methods in this task after incorporating pre-training BERT.

• Journal of Powder Materials
• /
• v.28 no.6
• /
• pp.455-461
• /
• 2021

In this study, multilayered SnO nanoparticles are prepared using oleylamine as a surfactant at 165℃. The physical and chemical properties of the multilayered SnO nanoparticles are determined by transmission electron microscopy (TEM), scanning electron microscopy (SEM), energy-dispersive spectroscopy (EDS), X-ray diffraction (XRD), and X-ray photoelectron spectroscopy (XPS). Interestingly, when the multilayered SnO nanoparticles are heated at 400℃ under argon for 2 h, they become more efficient anode materials, maintaining their morphology. Heat treatment of the multilayered SnO nanoparticles results in enhanced discharge capacities of up to 584 mAh/g in 70 cycles and cycle stability. These materials exhibit better coulombic efficiencies. Therefore, we believe that the heat treatment of multilayered SnO nanoparticles is a suitable approach to enable their application as anode materials for lithium-ion batteries.

• Nuclear Engineering and Technology
• /
• v.54 no.7
• /
• pp.2670-2689
• /
• 2022

STREAM - a lattice transport calculation code with method of characteristics for the purpose of light water reactor analysis - has been developed by the Computational Reactor Physics and Experiment laboratory (CORE) of the Ulsan National Institute of Science and Technology (UNIST). Recently, efforts have been taken to develop a photon module in STREAM to assess photon heating and the influence of gamma photon transport on power distributions, as only neutron transport was considered in previous STREAM versions. A multi-group photon library is produced for STREAM based on the ENDF/B-VII.1 library with the use of the library-processing code NJOY. The developed photon solver for the computation of 2D and 3D distributions of photon flux and energy deposition is based on the method of characteristics like the neutron solver. The photon library and photon module produced and implemented for STREAM are verified on VERA pin and assembly problems by comparison with the Monte Carlo code MCS - also developed at UNIST. A short analysis of the impact of photon transport during depletion and thermal hydraulics feedback is presented for a 2D core also from the VERA benchmark.

• Wind and Structures
• /
• v.32 no.5
• /
• pp.487-499
• /
• 2021

The wind-blown sand effect on the high-speed train is investigated. Unsteady RANS equation and the SST k-ω turbulent model coupled with the discrete phase model (DPM) are utilized to simulate the two-phase of air-sand. Sand impact force is calculated based on the Hertzian impact theory. The different cases, including various wind velocity, train speed, sand particle diameter, were simulated. The train's flow field characteristics and the sand impact force were analyzed. The results show that the sand environment makes the pressure increase under different wind velocity and train speed situations. Sand impact force increases with the increasing train speed and sand particle diameter under the same particle mass flow rate. The train aerodynamic force connected with sand impact force when the train running in the wind-sand environment were compared with the aerodynamic force when the train running in the pure wind environment. The results show that the head car longitudinal force increase with wind speed increasing. When the crosswind speed is larger than 35m/s, the effect of the wind- sand environment on the train increases obviously. The longitudinal force of head car increases 23% and lateral force of tail increases 12% comparing to the pure wind environment. The sand concentration in air is the most important factor which influences the sand impact force on the train.

• International Journal of Computer Science & Network Security
• /
• v.22 no.3
• /
• pp.220-228
• /
• 2022

The directed tests produce an expectation model to assist the organization's heads and professionals with settling on the right and speedy choice. A directed deep learning strategy has been embraced and applied for SCADA information. In this paper, for the load shedding expectation overall power organization of Libya, a convolutional neural network with multi neurons is utilized. For contributions of the neural organization, eight convolutional layers are utilized. These boundaries are power age, temperature, stickiness and wind speed. The gathered information from the SCADA data set were pre-handled to be ready in a reasonable arrangement to be taken care of to the deep learning. A bunch of analyses has been directed on this information to get a forecast model. The created model was assessed as far as precision and decrease of misfortune. It tends to be presumed that the acquired outcomes are promising and empowering. For assessment of the outcomes four boundary, MSE, RMSE, MAPE and R2 are determined. The best R2 esteem is gotten for 1-overlap and it was 0.98.34 for train information and for test information is acquired 0.96. Additionally for train information the RMSE esteem in 1-overlap is superior to different Folds and this worth was 0.018.

• Structural Engineering and Mechanics
• /
• v.85 no.6
• /
• pp.765-780
• /
• 2023

In this work, a novel combined logarithmic, secant and tangential 2D and quasi-3D refined higher order shear deformation theory is proposed to examine the buckling analysis of simply supported uniform functionally graded plates under uniaxial and biaxial loading. The proposed formulations contain a reduced number of variables compared to others similar solutions. The combined function employed in this study ensures automatically the zero-transverse shear stresses at the free surfaces of the structure. Various models of the material distributions are considered (linear, quadratic, cubic inverse quadratic and power-law). The differentials stability equations are derived via virtual work principle with including the stretching effect. The Navier's approach is applied to solve the governing equations which satisfying the boundary conditions. Several comparative and parametric studies are performed to illustrates the validity and efficacity of the proposed model and the various factors influencing the critical buckling load of thick FG plate.

• ETRI Journal
• /
• v.45 no.1
• /
• pp.28-44
• /
• 2023

We propose short packet communication in an underlay cognitive radio network assisted by an intelligent reflecting surface (IRS) composed of multiple reconfigurable reflectors. This scheme, called the IRS protocol, operates in only one time slot (TS) using the IRS. The IRS adjusts its phases to give zero received cumulative phase at the secondary destination, thereby enhancing the end-to-end signal-to-noise ratio. The transmitting power of the secondary source is optimized to simultaneously satisfy the multi-interference constraints, hardware limitations, and performance improvement. Simulation and analysis results of the average block error rates (BLERs) show that the performance can be enhanced by installing more reconfigurable reflectors, increasing the blocklength, lowering the number of required primary receivers, or sending fewer information bits. Moreover, the proposed IRS protocol always outperforms underlay relaying protocols using two TSs for data transmission, and achieves the best average BLER at identical transmission distances between the secondary source and secondary destination. The theoretical analyses are confirmed by Monte Carlo simulations.

• Nuclear Engineering and Technology
• /
• v.54 no.12
• /
• pp.4460-4473
• /
• 2022

The behaviour of the fission gas plays an important role in the fuel rod performance. In a previous work, we presented a physics-based model describing intra- and inter-granular behaviour of radioactive fission gas. The model was implemented in SCIANTIX, a mesoscale module for fission gas behaviour, and assessed against the CONTACT 1 irradiation experiment. In this work, we present the multi-scale coupling between the TRANSURANUS fuel performance code and SCIANTIX, used as mechanistic module for stable and radioactive fission gas behaviour. We exploit the coupled code version to reproduce two integral irradiation experiments involving standard fuel rod segments in steady-state operation (CONTACT 1) and during successive power transients (HATAC C2). The simulation results demonstrate the predictive capabilities of the code coupling and contribute to the integral validation of the models implemented in SCIANTIX.

• Journal of Ceramic Processing Research
• /
• v.13 no.spc1
• /
• pp.75-77
• /
• 2012

The AlGaN-based vertical light-emitting diodes (LEDs) on thick GaN epilayer were fabricated by a hydride vapor phase epitaxy with multi sliding boat system. The optical and electrical characteristics of AlGaN-based vertical LEDs were evaluated using a scanning electron microscopy, electroluminescence and I-V measurements. The AlGaN-based vertical LEDs structure has hexagonal symmetry, 500 ㎛ in diameter and above 67 ㎛ in growth thickness. At the room-temperature, the broaded strong peak and relatively high intensity peak were gradually measured at 405 nm with increasing injection current. And a forward operator voltage was measured to be about 7.5 V.

• Nuclear Engineering and Technology
• /
• v.55 no.5
• /
• pp.1775-1782
• /
• 2023

Several reactor physics commissioning experiments were conducted to obtain the neutronic parameters at the beginning of the G.A. Siwabessy Multi-purpose Reactor (RSG-GAS) operation. These parameters are essential for the reactor to safety operate. Leveraging the experimental data, this study evaluated the calculated core reactivity, control rod reactivity worth, integral control rod reactivity curve, and fuel reactivity. Calculations were carried out with Serpent 2 code using the latest neutron cross-section data ENDF/B-VIII.0. The criticality calculations were carried out for the RSG-GAS first core up to the third core configuration, which has been done experimentally during these commissioning periods. The excess reactivity for the second and third cores showed a difference of 510.97 pcm and 253.23 pcm to the experiment data. The calculated integral reactivity of the control rod has an error of less than 1.0% compared to the experimental data. The calculated fuel reactivity value is consistent with the measured data, with a maximum error of 2.12%. Therefore, it can be concluded that the RSG-GAS reactor core model is in good agreement to reproduce excess reactivity, control rod worth, and fuel element reactivity.