• 한국가스학회지
• /
• 제23권5호
• /
• pp.52-58
• /
• 2019

본 연구에서는 국내 도시가스 인수량에 대한 예측 모델을 개발하였다. 국내의 도시가스 회사는 KOGAS에 차년도 수요를 예측하여 보고해야 하므로 도시가스 인수량 예측은 도시가스 회사에 중요한 사안이다. 도시가스 사용량에 영향을 미치는 요인은 용도구분에 따라 다소 상이하나, 인수량 데이터는 용도별 구분이 어렵기 때문에 특정 용도에 관계없이 영향을 주는 요인으로 외기온도를 고려하여 모델개발을 실시하였다.실험 및 검증은 JB주식회사의 2008년부터 2018년까지 총 11년 치 도시가스 인수량 데이터를 사용하였으며, 전통적인 시계열 분석 중 하나인 ARIMA(Auto-Regressive Integrated Moving Average)와 딥러닝 기법인 LSTM(Long Short-Term Memory)을 이용하여 각각 예측 모델을 구축하고 두 방법의 단점을 최소화하기 위하여 다양한 앙상블(Ensemble) 기법을 사용하였다. 본 연구에서 제안한 일별 예측의 오차율 절댓값 평균은 Ensemble LSTM 기준 0.48%, 월별 예측의 오차율 절댓값 평균은 2.46%, 1년 예측의 오차율 절댓값 평균은 5.24%임을 확인하였다.

• 한국물환경학회지
• /
• 제39권1호
• /
• pp.46-60
• /
• 2023

Climate change causes fluctuations in water quality in the aquatic environment, which can cause changes in water circulation patterns and severe adverse effects on aquatic ecosystems in the future. Therefore, research is needed to predict and respond to water quality changes caused by climate change in advance. In this study, we tried to predict the dissolved oxygen (DO), chlorophyll-a, and turbidity of the Paldang reservoir for about two weeks using long short-term memory (LSTM) and gated recurrent units (GRU), which are deep learning algorithms based on recurrent neural networks. The model was built based on real-time water quality data and meteorological data. The observation period was set from July to September in the summer of 2021 (Period 1) and from March to May in the spring of 2022 (Period 2). We tried to select an algorithm with optimal predictive power for each water quality parameter. In addition, to improve the predictive power of the model, an important variable extraction technique using random forest was used to select only the important variables as input variables. In both Periods 1 and 2, the predictive power after extracting important variables was further improved. Except for DO in Period 2, GRU was selected as the best model in all water quality parameters. This methodology can be useful for preventive water quality management by identifying the variability of water quality in advance and predicting water quality in a short period.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2021년도 춘계학술발표대회
• /
• pp.463-465
• /
• 2021

센서 데이터는 여러가지 원인으로 인해 데이터 결측치가 발생할 수 있으며, 결측치로 인한 데이터의 처리 방식에 따라 데이터 분석 결과가 다르게 해석될 수 있다. 이는 펫 헬스 케어 서비스에서 치명적인 문제로 연결될 수 있다. 따라서 본 논문에서는 펫 웨어러블 디바이스로부터 수집되는 다양한 센서 데이터의 결측치를 처리하기 위해 GATs(Graph Attention neTworks)와 LSTM(Long Short Term Memory)을 결합하여 활용한 데이터 결측치 처리 기법을 제안한다. 펫 웨어러블 디바이스의 센서 데이터가 서로 연관성을 가지고 있다는 점을 바탕으로 인접 노드의 Attention 수치와 Feature map을 도출한다. 이후 Prediction Layer 를 통해 결측치의 Feature 를 예측한다. 예측된 Feature 를 기반으로 Decoding 과정과 함께 결측치 보간이 이루어진다. 제안된 기법은 모델의 변형을 통해 이상치 탐지에도 활용할 수 있을 것으로 기대한다.

• International journal of advanced smart convergence
• /
• 제13권1호
• /
• pp.37-47
• /
• 2024

This study presents the results of designing a system that determines the location of a person in an indoor environment based on a single IMU sensor attached to the tip of a person's shoe in an area where GPS signals are inaccessible. By adjusting for human footfall, it is possible to accurately determine human location and trajectory by correcting errors originating from the Inertial Measurement Unit (IMU) combined with advanced machine learning algorithms. Although there are various techniques to identify stepping, our study successfully recognized stepping with 98.7% accuracy using an artificial intelligence model known as Long Short-Term Memory (LSTM). Drawing upon the enhancements in our methodology, this article demonstrates a novel technique for generating a 200-meter trajectory, achieving a level of precision marked by a 2.1% error margin. Indoor pedestrian navigation systems, relying on inertial measurement units attached to the feet, have shown encouraging outcomes.

• Nuclear Engineering and Technology
• /
• 제56권6호
• /
• pp.2266-2273
• /
• 2024

A machine learning platform is proposed for the diagnosis of a severe accident progression in a nuclear power plant. To predict the key parameters for accident management including lost signals, a long short term memory (LSTM) network is proposed, where multiple accident scenarios are used for training. Training and test data were produced by MELCOR simulation of the Fukushima Daiichi Nuclear Power Plant (FDNPP) accident at unit 3. Feature variables were selected among plant parameters, where the importance ranking was determined by a recursive feature elimination technique using RandomForestRegressor. To answer the question of whether a reduced order ML model could predict the complex transient response, we performed a systematic sensitivity study for the choices of target variables, the combination of training and test data, the number of feature variables, and the number of neurons to evaluate the performance of the proposed ML platform. The number of sensitivity cases was chosen to guarantee a 95 % tolerance limit with a 95 % confidence level based on Wilks' formula to quantify the uncertainty of predictions. The results of investigations indicate that the proposed ML platform consistently predicts the target variable. The median and mean predictions were close to the true value.

• Nuclear Engineering and Technology
• /
• 제56권5호
• /
• pp.1725-1732
• /
• 2024

In nuclear power plants, reactor coolant leakage can occur due to various reasons. Early detection of leaks is crucial for maintaining the safety of nuclear power plants. Currently, a detection system is being developed in Korea to identify reactor coolant system (RCS) leakage of less than 0.5 gpm. Typically, RCS leaks are detected by monitoring temperature, humidity, and radioactivity in the containment, and a water level in the sump. However, detecting small leaks proves challenging because the resulting changes in the containment humidity and temperature, and the sump water level are minimal. To address these issues and improve leak detection speed, it is necessary to quantify the leaks and develop an artificial intelligence-based leak detection system. In this study, we employed bidirectional long short-term memory, which are types of neural networks used in artificial intelligence, to predict the relative humidity in the leakage area for leak quantification. Additionally, an optimization technique was implemented to reduce learning time and enhance prediction performance. Through evaluation of the developed artificial intelligence model's prediction accuracy, we expect it to be valuable for future leak detection systems by accurately predicting the relative humidity in a leakage area.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제12권4호
• /
• pp.149-158
• /
• 2023

최근 데이터 활용이 중요해짐에 따라 데이터 센터의 중요도도 함께 높아지고 있다. 하지만 데이터 센터는 막대한 전력을 소모함과 동시에 24시간 가동되는 시설이기 때문에 환경적, 경제적 측면에서 문제가 되고 있다. 최근 딥러닝 기법들을 사용하여 트래픽을 예측하거나, 데이터 센터나 서버에서 사용되는 전력을 줄이는 연구들이 다양한 관점에서 이루어지고 있다. 그러나 서버에서 처리되는 트래픽 데이터양은 변칙적이며 이는 서버를 관리하기 어렵게 만든다. 또한, 서버 상황에 따라 서버를 가변적으로 관리하는 기법에 대한 연구들이 여전히 많이 요구되고 있다. 따라서 본 논문에서는 이러한 문제점을 해결하기 위해 시계열 데이터 예측에 강세를 보이는 장단기 기억 신경망 (Long-Term Short Memory, LSTM)을 기반으로 한 가변적인 서버 관리 기법을 제안한다. 제안된 모델을 통해 서버에서 사용되는 전력을 보다 효과적으로 줄일 수 있게 되며, 현업환경에서 이전보다 안정적이고 효율적으로 서버를 관리할 수 있게 된다. 제안된 모델의 검증을 위해 위키피디아 (Wikipedia)의 데이터 센터 중 6개의 데이터 센터의 전송 및 수신 트래픽 데이터를 수집한 뒤 통계기반 분석을 통해 각 트래픽 데이터의 관계를 분석 및 실험을 수행하였다. 실험 결과 본 논문에서 제안된 모델의 유의미한 성능을 통계적으로 검증하였으며 서버 관리를 안정적이고 효율적으로 수행할 수 있음을 보여주었다.

• ETRI Journal
• /
• 제45권4호
• /
• pp.650-665
• /
• 2023

A novel smart metering technique capable of anomaly detection was proposed for real-time home power management system. Smart meter data generated in real-time were obtained from 900 households of single apartments. To detect outliers and missing values in smart meter data, a deep learning model, the autoencoder, consisting of a graph convolutional network and bidirectional long short-term memory network, was applied to the smart metering technique. Power management based on the smart metering technique was executed by multi-objective optimization in the presence of a battery storage system and an electric vehicle. The results of the power management employing the proposed smart metering technique indicate a reduction in electricity cost and amount of power supplied by the grid compared to the results of power management without anomaly detection.

• International journal of advanced smart convergence
• /
• 제8권2호
• /
• pp.132-139
• /
• 2019

We address the problem about forecasting the direction of stock price movement in the Korea market. Recently, the deep neural network is popularly applied in this area of research. In deep neural network systems, proper parameter initialization reduces training time and improves the performance of the model. Therefore, in our study, we propose a novel parameter initialization technique and apply this technique for the stock price movement prediction model. Specifically, we design a framework which consists of two models: a base model and a main prediction model. The base model constructed with LSTM is trained by using the large data which is generated by a large amount of the stock data to achieve optimal parameters. The main prediction model with the same architecture as the base model uses the optimal parameter initialization. Thus, the main prediction model is trained by only using the data of the given stock. Moreover, the stock price movements can be affected by other related information in the stock market. For this reason, we conducted our research with two types of inputs. The first type is the stock features, and the second type is a combination of the stock features and the Korea Composite Stock Price Index (KOSPI) features. Empirical results conducted on the top five stocks in the KOSPI list in terms of market capitalization indicate that our approaches achieve better predictive accuracy and F1-score comparing to other baseline models.

• 한국멀티미디어학회논문지
• /
• 제20권5호
• /
• pp.769-781
• /
• 2017

Deep learning networks like Convolutional Neural Networks (CNNs) show successful performances in many computer vision applications such as image classification, object detection, and so on. For implementation of deep learning networks in embedded system with limited processing power and memory, deep learning network may need to be simplified. However, simplified deep learning network cannot learn every possible scene. One realistic strategy for embedded deep learning network is to construct a simplified deep learning network model optimized for the scene images of the installation place. Then, automatic training will be necessitated for commercialization. In this paper, as an intermediate step toward automatic training under fisheye camera environments, we study more precise human localization in fisheye images, and propose an accurate human localization method, Automatic Ground-Truth Labelling Method (AGTLM). AGTLM first localizes candidate human object bounding boxes by utilizing GoogLeNet-LSTM approach, and after reassurance process by GoogLeNet-based CNN network, finally refines them more correctly and precisely(tightly) by applying saliency object detection technique. The performance improvement of the proposed human localization method, AGTLM with respect to accuracy and tightness is shown through several experiments.