• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.811-813
• /
• 1996

High impedance fault can not be easily detected by conventional method. But if it would not be detected and cleared quickly, it can result in fires, and electric shock. In this paper, neural network, which has learning capability, is used for high impedance fault detector. The potential of the neural network approach is demonstrated by simulation using KEPCO's measured data. The instantaneous values and frequency spectrum of current are respectively used as the inputs of neural networks. Also, the methods using combined data to exploit the advantage of each data are proposed. In this paper, back-propagation network(BPN) is used for high impedance fault detector and can use for high speed relay because it detects faults within 1 cycle.

• Journal of Electrical Engineering and Technology
• /
• v.9 no.5
• /
• pp.1482-1488
• /
• 2014

The reliability evaluation of the electrical distribution networks (EDN) requires sufficient consideration of the effects of planned outages. The planned outages of the EDN can be divided, by outage models and their effects on the reliability into two major categories: by equipment and by feeder. After studying the characteristics of different categories of planned outages, this paper expands the classification of load points by outage time from 4 types to 7 types and defines corresponding reliability parameters for the different types. By using the section algorithm, this paper proposes a reliability evaluation technique of EDN considering equipment random failures and two categories of planned outages. The proposed technique has been applied to the RBTS-BUS6 test system and some practical EDNs in China. The study results demonstrate that the proposed technique is of higher practical value and can be used for evaluating the reliability performance of EDN more efficiently considering the planned outages.

• Nuclear Engineering and Technology
• /
• v.53 no.2
• /
• pp.657-665
• /
• 2021

Nondestructive evaluation methods play an important role in ensuring component integrity and safety in many industries. Operator fatigue can play a critical role in the reliability of such methods. This is important for inspecting high value assets or assets with a high consequence of failure, such as aerospace and nuclear components. Recent advances in convolution neural networks can support and automate these inspection efforts. This paper proposes using residual neural networks (ResNets) for real-time detection of corrosion, including iron oxide discoloration, pitting and stress corrosion cracking, in dry storage stainless steel canisters housing used nuclear fuel. The proposed approach crops nuclear canister images into smaller tiles, trains a ResNet on these tiles, and classifies images as corroded or intact using the per-image count of tiles predicted as corroded by the ResNet. The results demonstrate that such a deep learning approach allows to detect the locus of corrosion via smaller tiles, and at the same time to infer with high accuracy whether an image comes from a corroded canister. Thereby, the proposed approach holds promise to automate and speed up nuclear fuel canister inspections, to minimize inspection costs, and to partially replace human-conducted onsite inspections, thus reducing radiation doses to personnel.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.54 no.7
• /
• pp.358-365
• /
• 2005

Transmission networks play an important role which transfer generated electric power to a consumer in power system operation. In a competitive environment of electric power industry, developing the technological criterions and methodologies on transmission planning is becoming new challenge to transmission system planner. The use of a locational signal and the provision of a indicative plan to control the transmission investment reasonably is very important in the viewpoint of a regulator. The main target of this study is to develop a systematic criterion of transmission expansion planning. And system congestion cost is considered. The proposed methodology was demonstrated with several case studies.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.10d
• /
• pp.372-375
• /
• 2007

본 논문에서는 이동 노드를 가지는 센서 네트워크와 같은 대규모 애드 혹 네트워크에서 사용 가능한 새로운 라우팅 프로토콜을 소개한다. 본 라우팅 프로토콜은 노드의 이동성을 효율적으로 지원하기 위해 GPS를 이용한 위치정보를 이용하며 가상 싱크 기법을 사용하여 위치정보의 빈번한 플러딩 없이 라우팅을 수행할 수 있도록 설계되었다. V-Geocast는 NS-2 시뮬레이터 환경에서 구현하였으며 단순한 Geocast나 AODV와 같은 기존의 애드 혹 라우팅 알고리즘에 대비하여 뛰어난 성능을 가지는 것을 확인하였다.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.15 no.1
• /
• pp.99-115
• /
• 1990

This paper is concerned with the unit commitment problem in an electric power system with both thermal and pumped-storage hydroelectric units. This is a mixed integer programming problem and the Lagrangean relaxation method is used. We show that the relaxed problem decomposes into two kinds of subproblems : a shortest-path problem for each thermal unit and a minimum cost flow problem for each pumped-storage hydroelectric unit. A method of obtaining an incumbenet solution from the solution of a relaxed problem is presented. The Lagrangean multipliers are updated using both subgradient and incremental cost. The algorithm is applied to a real Korean power generation system and its computational results are reported and compaired with other works.

• Journal of IKEEE
• /
• v.2 no.2 s.3
• /
• pp.255-262
• /
• 1998

The neural network can roughly be classified as the specialized control, indirect control and general schemes. Neural network is adopted for MPPT of solar array. And back propagation algorithm also is used to train neural network controller. We investigate the possibilities of $P_{max}$ control using the neural networks, and then we also examine about operating the solar cell at an optimal voltage comprise of temperature compensated voltage with boost converter. Proposed boost converter of MPPT system is studied by simulation and is implemented by using a microprocessor(80c196kc) which controls duty ratio of the boost converter.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.32A no.4
• /
• pp.47-56
• /
• 1995

In this paper the model of the MQE-IMD-based neural trigger circuit is improved, where MQW-IMD is a new semiconductor device proposed and experimentally demonstrated by the author for the hardware implementation of the neural networks. The electron energy of AlXGa1-XAsbarrier is calculated by Ensemble Monte Carlo simulation according to the variation of Al mole fraction x and the applied electric field, whtich had been roughly estimated in the previous paper because of the difficulty to get the data. And in the consideration of the tunneling of the confined electrons within the quantum well the accuracy of the impact ionization rate is enhaned. Finally, the dependance of the frequency of pulse-train on the number of quantum wells can be calculated by modelling the effect of the distance of the induced positive charge from the cathode on the electric field at the cathode.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.61 no.1
• /
• pp.29-34
• /
• 2012

Due to the increasing of nonlinear loads such as converters and inverters connected to the electric power distribution system, and extensive application of harmonic generation sources with power electronic devices, disturbance of the electric power system and its influences on industries have been continuously increasing. Thus, it is difficult to construct accurate load model for active and reactive power in environments with harmonics. In this research, we develop a load modeling method based on Extreme Learning Machine(ELM) with fast learning procedure for residential loads. Using data sets acquired from various residential loads, the proposed method has been intensively tested. As the experimental results, we confirm that the proposed method makes it possible to effective estimate active and reactive powers than conventional methods.

• Proceedings of the Korean Society of Soil and Groundwater Environment Conference
• /
• 2006.04a
• /
• pp.391-393
• /
• 2006

In this study, the performance of artificial neural network (ANN) models for estimating spatial distribution of water quality was evaluated using electric conductivity (EC) values in landfill site. For the ANN model development, feedforward neural networks and backpropagation algorithm with gradient descent method were used. In Test 1, the interpolation ability of the ANN model was evaluated. Results of the ANN model were more precise than those of the Kriging model. In Test 2, spatial distributions of EC values were predicted using precipitation data. Results seemed to be reasonable, however, they showed a limitation of ANN models in extrapolations.