• Journal of international Conference on Electrical Machines and Systems
• /
• v.3 no.1
• /
• pp.48-53
• /
• 2014

In December 2011 and July 2012, two sets of 840MVA hydro-generator of Three Gorges on Yangtze River with Close-loop-self-circulating evaporative cooling (CLSCEC) system were put into commercial operation. In this paper, we make engineering summary of these two generators with CLSCEC system. We also make a comparison between the internal water cooling (IWC) hydro-generator and the CLSCEC hydro-generator used in Three Gorges power plant in fields of their operating characteristics, working performances, technical features, working safety and reliability. In addition, engineering structures, type tests' results and systematic emulating calculation of CLSCEC schemes are analyzed.

• Journal of the Optical Society of Korea
• /
• v.17 no.5
• /
• pp.357-361
• /
• 2013

A regenerative Er-doped fiber amplifier system for a high-repetition-rate optical pulse train is investigated for the first time. A signal pulse train with a wavelength tuning range of 18 nm is produced by a passive mode-locked fiber laser based on a nonlinear polarization rotation technique. In order to realize the amplification, an optical delay-line is used to achieve time match between the pulses' interval and the period of pulse running through the regenerative amplifier. The 16 dB gain is obtained for an input pulse train with a launching power of -30.4 dBm, a center wavelength of 1563.4 nm and a repetition rate of 15.3 MHz. The output properties of signal pulses with different center wavelengths are also discussed. The pulse amplification is found to be different from the regenerative amplification system for CW signals.

• Current Optics and Photonics
• /
• v.6 no.6
• /
• pp.627-633
• /
• 2022

To realize uniform side pumping of solar lasers and improve their output power, a solar concentrating system based on off-axis parabolic mirrors is proposed. Four identical off-axis parabolic mirrors with focal length of 1,000 mm are toroidally arranged as the primary concentrator. Four two-dimensional compound parabolic concentrators (2D-CPCs) are designed as a secondary concentrator to further compress the focused spot induced by the parabolic mirrors, and the focused light is then homogenized by four rectangular diffusers and provides uniform pumping for a laser-crystal rod to achieve solar laser emission. Simulation results show that the solar power received by the laser rod, uniformity of the light spot, and output power of the solar laser are 7,872.7 W, 98%, and 351.8 W respectively. This uniform pumping configuration and concentrator design thus provide a new means for developing high-power side-pumped solid-state solar lasers.

• Nuclear Engineering and Technology
• /
• v.56 no.4
• /
• pp.1284-1295
• /
• 2024

Due to increasing operational security demands, digital and intelligent condition monitoring of nuclear power plants is becoming more significant. However, establishing an accurate and effective anomaly detection model is still challenging. This is mainly because of data characteristics of nuclear power data, including the lack of clear class labels combined with frequent interference from outliers and anomalies. In this paper, we introduce a Transformer-based unsupervised model for anomaly detection of nuclear power data, a modified loss function based on the maximum correntropy criterion (MCC) is applied in the model training to improve the robustness. Experimental results on simulation datasets demonstrate that the proposed Trans-MCC model achieves equivalent or superior detection performance to the baseline models, and the use of the MCC loss function is proven can obviously alleviate the negative effect of outliers and anomalies in the training procedure, the F1 score is improved by up to 0.31 compared to Trans-MSE on a specific dataset. Further studies on genuine nuclear power data have verified the model's capability to detect anomalies at an earlier stage, which is significant to condition monitoring.

• Journal of Microbiology and Biotechnology
• /
• v.31 no.7
• /
• pp.921-932
• /
• 2021

LINC01272 is a long non-coding RNA (lncRNA) that has been considered as a biomarker for many diseases including lung squamous cell carcinoma. Here, we investigated the function and mechanism of LINC01272 on lung cancer (LC). The differential expression of LINC01272 in LC and normal samples was analyzed by GEPIA based on the data from TCGA-LUAD database, as survival prognosis was analyzed through Kaplan-Meier Plotter. LINC01272 overexpression plasmid and miR-7-5p mimic were transfected into A549 and PC-9 cells. LINC01272, miR-7-5p and cardiolipin synthase 1 (CRLS1) mRNA expression was measured by quantitative reverse transcription-polymerase chain reaction. Cell viability was detected through MTT assay. Cell multiplication was evaluated by cell formation assay. Cell apoptosis was assessed through flow cytometry assay. Through bioinformatics, the target miRNA of LINC01272 and downstream genes of miR-7-5p were predicted. The targeting relationship was tested by dual luciferase reporter analysis. CRLS1, B-cell lymphoma-2 (Bcl-2), BCL2-associated X (Bax) and cleaved caspase-3 protein levels were detected through western blot. LINC01272 was downregulated in LC and low LINC01272 expression had poor prognosis. In A549 and PC-9 cells, LINC01272 inhibited cell viability and multiplication and induced apoptosis. LINC01272 negatively regulated miR-7-5p and CRLS1 was a target of miR-7-5p. MiR-7-5p reversed the effect of LINC01272 on viability, multiplication, apoptosis and expression of miR-7-5p and CRLS1 as well as apoptosis-related factors (Bcl-2, Bax and cleaved caspase-3). LINC01272 suppressed cell multiplication and induced apoptosis via regulating the miR-7-5p/CRLS1 axis in LC.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1759-1772
• /
• 2023

Chinese named entity recognition (NER) is a challenging work that seeks to find, recognize and classify various types of information elements in unstructured text. Due to the Chinese text has no natural boundary like the spaces in the English text, Chinese named entity identification is much more difficult. At present, most deep learning based NER models are developed using a bidirectional long short-term memory network (BiLSTM), yet the performance still has some space to improve. To further improve their performance in Chinese NER tasks, we propose a new NER model, IDCNN-BiLSTM-Highway, which is a combination of the BiLSTM, the iterated dilated convolutional neural network (IDCNN) and the highway network. In our model, IDCNN is used to achieve multiscale context aggregation from a long sequence of words. Highway network is used to effectively connect different layers of networks, allowing information to pass through network layers smoothly without attenuation. Finally, the global optimum tag result is obtained by introducing conditional random field (CRF). The experimental results show that compared with other popular deep learning-based NER models, our model shows superior performance on two Chinese NER data sets: Resume and Yidu-S4k, The F1-scores are 94.98 and 77.59, respectively.

• Journal of Adhesion and Interface
• /
• v.7 no.4
• /
• pp.18-23
• /
• 2006

The influence of the notch length on the stress distribution of mid-bondline and adherend was investigated using elasto-plastic finite element method. The results from the simulation showed that peak stress of mid-bondline decreased markedly as adherend with notch in the middle of lap zone, and the stress in the middle of joint with low stress originally increased evidently. All the peak stresses decreased firstly and increased again as the length of notch increased. The relative higher peak stress appeared at the point near the notch of adherend where might be failed previously during the loading procedure.

• Structural Engineering and Mechanics
• /
• v.64 no.2
• /
• pp.271-278
• /
• 2017

In this paper, we propose a new conjugate gradient method which possesses the global convergence and apply it to solve inverse problems of the dynamic loads identification. Moreover, we strictly prove the stability and convergence of the proposed method. Two engineering numerical examples are presented to demonstrate the effectiveness and speediness of the present method which is superior to the Landweber iteration method. The results of numerical simulations indicate that the proposed method is stable and effective in solving the multi-source dynamic loads identification problems of practical engineering.

• Journal of applied mathematics & informatics
• /
• v.29 no.5_6
• /
• pp.1285-1293
• /
• 2011

We establishes the polynomial convergence of a new class of path-following methods for semidefinite linear complementarity problems (SDLCP) whose search directions belong to the class of directions introduced by Monteiro [9]. Namely, we show that the polynomial iteration-complexity bound of the well known algorithms for linear programming, namely the predictor-corrector algorithm of Mizuno and Ye, carry over to the context of SDLCP.

• Proceedings of the Korean Magnestics Society Conference
• /
• 2016.05a
• /
• pp.79-79
• /
• 2016