• Journal of the Korean Society of Systems Engineering
• /
• v.18 no.2
• /
• pp.75-93
• /
• 2022

Machine learning (ML) data-driven meta-model is proposed as a surrogate model to reduce the excessive computational cost of the physics-based model and facilitate the real-time prediction of a nuclear power plant's transient response. To forecast the transient response three machine learning (ML) meta-models based on recurrent neural networks (RNNs); specifically, Long Short Term Memory (LSTM), Gated Recurrent Unit (GRU), and a sequence combination of Convolutional Neural Network (CNN) and LSTM are developed. The chosen accident scenario is a control element assembly withdrawal at power concurrent with the Loss Of Offsite Power (LOOP). The transient response was obtained using the best estimate thermal hydraulics code, MARS-KS, and cross-validated against the Design and control document (DCD). DAKOTA software is loosely coupled with MARS-KS code via a python interface to perform the Best Estimate Plus Uncertainty Quantification (BEPU) analysis and generate a time series database of the system response to train, test and validate the ML meta-models. Key uncertain parameters identified as required by the CASU methodology were propagated using the non-parametric Monte-Carlo (MC) random propagation and Latin Hypercube Sampling technique until a statistically significant database (181 samples) as required by Wilk's fifth order is achieved with 95% probability and 95% confidence level. The three ML RNN models were built and optimized with the help of the Talos tool and demonstrated excellent performance in forecasting the most probable NPP transient response. This research was guided by the Systems Engineering (SE) approach for the systematic and efficient planning and execution of the research.

• Journal of the Korean Society of Systems Engineering
• /
• v.18 no.2
• /
• pp.94-107
• /
• 2022

Accidents prevention and mitigation is the highest priority of nuclear power plant (NPP) operation, particularly in the aftermath of the Fukushima Daiichi accident, which has reignited public anxieties and skepticism regarding nuclear energy usage. To deal with accident scenarios more effectively, operators must have ample and precise information about key safety parameters as well as their future trajectories. This work investigates the potential of machine learning in forecasting NPP response in real-time to provide an additional validation method and help reduce human error, especially in accident situations where operators are under a lot of stress. First, a base-case SGTR simulation is carried out by the best-estimate code RELAP5/MOD3.4 to confirm the validity of the model against results reported in the APR1400 Design Control Document (DCD). Then, uncertainty quantification is performed by coupling RELAP5/MOD3.4 and the statistical tool DAKOTA to generate a large enough dataset for the construction and training of neural-based machine learning (ML) models, namely LSTM, GRU, and hybrid CNN-LSTM. Finally, the accuracy and reliability of these models in forecasting system response are tested by their performance on fresh data. To facilitate and oversee the process of developing the ML models, a Systems Engineering (SE) methodology is used to ensure that the work is consistently in line with the originating mission statement and that the findings obtained at each subsequent phase are valid.

• Nuclear Engineering and Technology
• /
• v.55 no.9
• /
• pp.3423-3440
• /
• 2023

In this work, a multivariate time-series machine learning meta-model is developed to predict the transient response of a typical nuclear power plant (NPP) undergoing a steam generator tube rupture (SGTR). The model employs Recurrent Neural Networks (RNNs), including the Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and a hybrid CNN-LSTM model. To address the uncertainty inherent in such predictions, a Bayesian Neural Network (BNN) was implemented. The models were trained using a database generated by the Best Estimate Plus Uncertainty (BEPU) methodology; coupling the thermal hydraulics code, RELAP5/SCDAP/MOD3.4 to the statistical tool, DAKOTA, to predict the variation in system response under various operational and phenomenological uncertainties. The RNN models successfully captures the underlying characteristics of the data with reasonable accuracy, and the BNN-LSTM approach offers an additional layer of insight into the level of uncertainty associated with the predictions. The results demonstrate that LSTM outperforms GRU, while the hybrid CNN-LSTM model is computationally the most efficient. This study aims to gain a better understanding of the capabilities and limitations of machine learning models in the context of nuclear safety. By expanding the application of ML models to more severe accident scenarios, where operators are under extreme stress and prone to errors, ML models can provide valuable support and act as expert systems to assist in decision-making while minimizing the chances of human error.

• Journal of the Korean Geotechnical Society
• /
• v.22 no.12
• /
• pp.45-55
• /
• 2006

This paper presents the results of full-scale uplift load tests performed on 30 passive anchors grouted to various lengths at Taean site in Korea. Various rock types were tested, ranging from highly weathered to sound gneiss. Rock anchors were installed over a wide range of rock types and qualities with a fixed anchored depth of $1{\sim}4m$. The majority of installations used SD4O-D51 no high grade steel rebar to induce rock failure prior to rod failure. In many tests, rock failure was reached and the ultimate loads were recorded along with observations of the shape and extent of the failure surface. The test results, the failure mechanisms as well as uplift capacities of rock anchors depend mostly on rock type and quality, embedded fixed length, and the strength of rebar. Based on test results, the main parameters governing the uplift capacity of the rock anchor system were determined.

• Tunnel and Underground Space
• /
• v.33 no.5
• /
• pp.425-434
• /
• 2023

To measure the wear amount of the TBM disk cutter in real time, it is important not only to automate the measurement using sensors, but also to stably transmit the measured data to the information processing system. In this study, we investigated the viability of utilizing Bluetooth Low Energy (BLE) technology to wirelessly transmit sensor data from the TBM cutter head to a receiver located at the chamber's rear. Through laboratory experiments, we analyzed the Received Signal Strength Index (RSSI) of the receiver considering various signal strength of the transmitter, separation distances between the transmitter and receiver and chamber fill materials. Our results demonstrate that wireless data transmission is feasible across all tested conditions when the transmitter signal strength is 0 dBm or higher.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.52 no.7
• /
• pp.429-436
• /
• 2003

This paper presents a new approach for evaluating the transmission marginal loss factors (MLFs) considering the reactive power. Generally, MLFs are represented as the sensitivity of transmission losses, which is computed from the change of the generation at reference bus by the change of the load at the arbitrary bus-i. The conventional evaluation method for MLFs uses the only H matrix, which is a part of jacobian matrix. Therefore, the MLFs computed by the existing method, don't consider the effect of the reactive power, although the transmission losses are a function of the reactive power as well as the active power. To compensate the limits of the existing method for evaluating MLFs, the power factor at the bus-i is introduced for reflecting the effect of the reactive power in the evaluation method of the MLFs. Also, MLFs calculated by the developed method are applied to energy spot markets to reflect the impacts of reactive power. This method is tested with the sample system with 5-bus, and analyzed how much MLFs have an effect on the bidding/offer price, market clearing price(MCP), and settlement in the competitive energy spot market. This paper compared the results of MLFs calculated by the existing and proposed method for the IEEE 14-bus system, and the KEPCO system.

• 한국신재생에너지학회:학술대회논문집
• /
• 2008.05a
• /
• pp.229-234
• /
• 2008

Korea Electric Power Corporation (KEPCO) has started the nation's first biogas-microturbine project in the city of Gongju as an effort to encourage the utilization of wasted biogas containing useful energy source in the form of $CH_4$. The goal of the project is to set up the biogas microturbine co-generation system for utilizing biogas as an energy source and improving the economics of the wastewater treatment plant. Wastewater treatment processes were investigated in depth to find improvement possibility. Changes in internal recirculation ratio and pre-treatment degree are needed to optimize plant operation and biogas production. Biogas pre-treatment system satisfies Capstone's fuel condition requirement with the test result of 99.9% and 90.2% of hydrogen sulphide and ammonia is removal performance. Installation of microturbine and manufacture of heat exchanger to warm anaerobic digester has been done successfully. Expected economic profit produced by the system is coming from energy saving including electricity 115,871kWh/year and heat contained in exhaust gas 579GJ/year.

• Proceedings of the KIEE Conference
• /
• 2005.11b
• /
• pp.271-273
• /
• 2005

In KEPCO(Korea Electric Power Corporation)system, 40% of total loads is ones of metropolitan area and large staled generation plants are located out of metropolitan area. Therefore, to decide the maximum power transfer capability of interface line between the regions is essential for planning, control and operation of efficient transmission system. For this reason, prior study suggested calculating the limit of our interface line from the side of voltage stability. However, the presented plan had a limit partially to apply practice power system. Therefore this paper suggests a algorithm that change the generation output as the merit order instead of prior method that distribute according to proportion of present output. This method could be applied closer to practice than the existing algorithm because it consider generation cost.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.9 no.5
• /
• pp.1145-1149
• /
• 2008

This paper presents KEPCO research project for developing wide area DAS by reengineering conventional DAS and its functional upgrade. Through the research project system design requirements were prepared based on the local operation status and oversea examples. Application softwares were also developed supporting integrated operation of several area. The system implementation was performed on Choongnam regional head office for pilot operation. The developed wide area DAS will be fully installed on 41 DCC nation wide and expected to contribute on innovation of our distribution network operation.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.53 no.1
• /
• pp.7-12
• /
• 2004

The role of protective relay with correct setting value is very important for the reliable operation of power : system. The setting process of a protective relay is difficult and requires hard work of calculations. Therefore, KEPCO as been using PROSET2000, a computer program system, in order to automate the setting process of protective relays in power transmission network since 2000. This system has a database and relay setting modules for each type of realy with coresponding Human and Computer Interaction(HCI). For the setting of a new type of relay, it require a new module, a HCI and a database table corresponding to the new relay. But it's update is not easy in case of new type of relay. This paper discussed the update method introducing a concept of Open Interface. It addressd a method to make a new HCI for new relay in view of object-oriented method using UML(Unified Modeling Language). Using the UML, we updated the PROSET2000. HCI screens for a new relay is generated automatically using the information of the new relay stored in the database. And, the database management module is upgraded to solve problems in manage and maintain the database in view of Open Interface.