• Nuclear Engineering and Technology
• /
• v.56 no.9
• /
• pp.3740-3749
• /
• 2024

To reduce the losses caused by aging failure of insulation gate bipolar transistor (IGBT), which is the core components of nuclear power plant rod position indicating and rod control (RPC) system. It is necessary to conduct studies on its life prediction. The selection of IGBT failure characteristic parameters in existing research relies heavily on failure principles and expert experience. Moreover, the analysis and learning of time-domain degradation data have not been fully conducted, resulting in low prediction efficiency as the monotonicity, time correlation, and poor anti-interference ability of extracted degradation features. This paper utilizes the advantages of the stacked denoising autoencoder(SDAE) network in adaptive feature extraction and denoising capabilities to perform adaptive feature extraction on IGBT time-domain degradation data; establishes a long-short-term memory (LSTM) prediction model, and optimizes the learning rate, number of nodes in the hidden layer, and number of hidden layers using the Gray Wolf Optimization (GWO) algorithm; conducts verification experiments on the IGBT accelerated aging dataset provided by NASA PCoE Research Center, and selects performance evaluation indicators to compare and analyze the prediction results of the SDAE-LSTM model, PSOLSTM model, and BP model. The results show that the SDAE-LSTM model can achieve more accurate and stable IGBT life prediction.

• Korean Journal of Adult Nursing
• /
• v.24 no.2
• /
• pp.149-159
• /
• 2012

Purpose: The purpose of this study was to understand the life of newly appointed professors of nursing by exploring the meaning and essence of their academic work and life experience. Methods: The participants consisted of 11 newly appointed female full-time professors with less than five years of experience who had been purposely sampled from three or four-year nursing colleges. Data were collected from October 2010 to December 2011 through the face-to-face interview and phone interview. Thematic analysis was conducted using hermeneutic phenomenological methodology. Results: The findings were classified into seven themes: 'Starting feeling hope and delight', 'putting sustained energy into lecturing and training', 'struggling in the tense atmosphere where assessments are often conducted', 'gaining courage in an associated relationship network', 'experiencing confusion of the identity as a professor', 'seeking compromise, handling their work and housework', and 'shedding new light on reality with self-awareness'. Conclusion: This study is significant in that it offered the foundation for the creation of human resources networks and mentoring programs that help newly appointed professors to effectively adjust to university society on the basis of the sympathetic understanding of and support for their life.

• Proceedings of the Korean Society of Applied Pharmacology
• /
• 2002.07a
• /
• pp.73-78
• /
• 2002

Microfluidic lab-on-a-chip (LOC) systems have enabled a new generation ofassay technologies in chemical and biomedical sciences. Caliper's microfluidic LOC systems contain a network of microscopic channels through which fluids and chemical are moved in order to perform experiments. The main advantages of these continuous-flow devices are integration and automation of multiple steps in complex analytical procedures to improve the reproducibility of the results, and eliminated the manual labor, time and pipetting errors involved in analyses. The present talk is devoted to give a brief introduction of microfluidic basics and to present in applying continuous-flow microchips to drug screening with model enzyme assays.

• Proceedings of the Korean Institute of Surface Engineering Conference
• /
• 2009.10a
• /
• pp.233-234
• /
• 2009

증착된 silicon nitride (Sin) 박막의 수명 시간을 예측하는 신경망 모델을 개발하였다. SiN 박막은 플라즈마 화학기상 증착방식을 이용하여 증착되었다. 증착 공정은 통계적인 실험계획표를 이용하여 수행되었고, 신경망 모델의 예측 성능은 유전자 알고리즘을 이용하여 최적화하였다. 수명시간은 다른 박막특성 (굴절률, 증착률, 전하밀도)의 영향을 상당히 받았으며, 특히 굴절률과 전하밀도는 높은 증착률에서 증가시킬 때 수명시간을 최대화할 수 있었다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.5
• /
• pp.380-384
• /
• 2009

The log data collected on a mobile device contain diverse and continuous information about the user. From the log data, the location, pictures, running functions and services of the user can be obtained. It has interested in the research inferring the contexts and understanding the everyday-life of mobile users. In this paper, we have studied the methods for real-time collection of log data from mobile devices, analysis of the data, map based visualization and effective management of the personal everyday-life information. We have developed an application for sharing the contexts. The proposed application infers the personal contexts with Bayesian network probabilistic model. In the experiments, we confirm that the usability of visualization and information sharing functions based on the real world log data.

• Journal of Korea Multimedia Society
• /
• v.12 no.9
• /
• pp.1297-1308
• /
• 2009

The life time of a wireless sensor network composed of numerous sensor nodes depend on ones of its sensor nodes. The energy efficiency operation of nodes, therefore, is one of the crucial factors to design the network. Researches based on the hierarchical network topology have been proposed and evolved in terms of energy efficiency. However, in existing researches, application layer data obtained from sensor nodes are not considered properly to compose cluster, including issue that nodes communicate with their cluster heads in TDMA scheduling. In this paper, we suggest an energy-efficient topology control scheme based on application layer data in wireless sensor networks. By using application layer data, sensor nodes form a section which is defined as the area of adjacent nodes that retain similar characteristics of application environments. These sections are further organized into clusters. We suggest an algorithm for selecting a cluster head as well as a way of scheduling to reduce the number of unnecessary transmissions from each node to its cluster head, which based on the degree and the duration of similarity between the node's data and its head's data in each cluster without seriously damaging the integrity of application data. The results show that the suggested scheme can save the energy of nodes and increase the life time of the entire network.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.11
• /
• pp.5227-5237
• /
• 2011

Technology roadmap is a powerful tool that considers relationships of technology, product and market and referred as a supporting technology strategy and planning. There are numerous studies that have attempted to develop technology roadmap and case studies on specific technology areas. However, a number of studies have been dependant on brainstorming and discussion of expert group, delphi technique as qualitative analysis rather than systemic and quantitative analysis. To overcome the limitation, patent analysis considered as quite quantitative analysis is employed in this paper. Therefore, this paper proposes new technology roadmapping based on patent citation network considering technology life cycle and suggests planning for undeveloped technology but considered as promising. At first, patent data and citation information are collected and patent citation network is developed on the basis of collected patent information. Secondly, we investigate a stage of technology in the life cycle by considering patent application year and the technology life cycle, and duration of technology development is estimated. In addition, subsequent technologies are grouped as nodes of a super-level technology to show the evolution of the technology for the period. Finally, a technology roadmap is drawn by linking these technology nodes in a technology layer and estimating the duration of development time. Based on technology roadmap, technology planning is conducted to identify undeveloped technology through text mining and this paper suggests characteristics of technology that needs to be developed in the future. In order to illustrate the process of the proposed approach, technology for hydrogen storage is selected in this paper.

• Proceedings of the Korea Society of Information Technology Applications Conference
• /
• 2005.11a
• /
• pp.27-30
• /
• 2005

Because of unreplenishable power resources, reducing node energy consumption to extend network lifetime is an important requirement in wireless sensor networks. In addition both path length and path cost are important metrics affecting sensor lifetime. We propose a dynamic clustering scheme based on location in wireless sensor networks. Our scheme can localize the effects of route failures, reduce control traffic overhead, and thus enhance the reachability to the destination. We have evaluated the performance of our clustering scheme through a simulation and analysis. We provide simulation results showing a good performance in terms of approximation ratios.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06d
• /
• pp.325-327
• /
• 2012

배터리를 기반으로 동작하는 무선센서네트워크 환경에서는 전력이 가장 중요한 자원이다. 본고는 무선센서네트워크 환경에서 무선전력전송을 이용하여 센서 네트워크의 생명(Network Life Time)을 최대화하는 네트워크 환경을 제안하였다. 또한 전력을 전송하는 각 소스가 네트워크 생명을 최대화하기 위해 스케줄링을 결정짓는 메트릭을 제시하였다. 제안 모델은 지향성 안테나가 가지는 전송 범위의 구역을 나누고 각 영역에 대해서 그 영역의 전력 수급, 배터리 소모량, 구역의 면적당 센서 수를 이용하여 효과적으로 전력을 전달할 수 있는 메트릭을 도출하였다.

• The Transactions of the Korean Institute of Electrical Engineers B
• /
• v.50 no.8
• /
• pp.416-426
• /
• 2001

This paper deals with the impact of an inverter station on the torsional dynamics of turbine-generator which is located at the inverter side of a HVDC-AC network power system. The studies show that the torsional stress of turbine-generator depends on the AC network fault locations because of the commutation failures of inverter station. And the torsional stress induce fatigue in the shaft material and reduce the shaft life-time. So, the purpose of this paper is to analysis the torsional stress of turbine-generator shaft at inverter side, to find the checked points of turbine-generator. The EMTDC program is used for the simulation studies.