• Journal of Broadcast Engineering
• /
• v.24 no.5
• /
• pp.870-878
• /
• 2019

In this paper, we propose an Attention-based BLSTM for predicting the video compression standard of a video. Recently, in NLP, many researches have been studied to predict the next word of sentences, classify and translate sentences by their semantics using the structure of RNN, and they were commercialized as chatbots, AI speakers and translator applications, etc. LSTM is designed to solve the gradient vanishing problem in RNN, and is used in NLP. The proposed algorithm makes video compression standard prediction possible by applying BLSTM and Attention algorithm which focuses on the most important word in a sentence to a bitstream of a video, not an sentence of a natural language.

• Journal of Ocean Engineering and Technology
• /
• v.33 no.1
• /
• pp.33-41
• /
• 2019

Estimating fatigue damage is a very important issue in the design of ships. The springing and whipping response, which is the hydro-elastic response of the ship, can increase the fatigue damage of the ship. So, these phenomena should be considered in the design stage. However, the current studies on the the application of springing and whipping responses at the design stage are not sufficient. So, in this study, a prediction method was developed using fluid-structural interaction analysis to assess of the fatigue damage induced by springing and whipping. The stress transfer function (Stress RAO) was obtained by using the 3D FE model in the frequency domain, and the fatigue damage, including linear springing, was estimated by using the wide band damage model. We also used the 1D beam model to develop a method to estimate the fatigue damage, including nonlinear springing and whipping by the vertical bending moment in the short-term sea state. This method can be applied to structural members where fatigue strength is weak to vertical bending moments, such as longitudinal stiffeners. The methodology we developed was applied to 325K VLOC, and we analyzed the effect of the springing and whipping phenomena on the existing design.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.411-411
• /
• 2015

유량예측의 가장 주된 목적은 가뭄과 홍수와 같은 수해방지를 위해 통합수자원관리를 수행하는데 있다. 이런 유량예측을 위해 다양한 기법들로 예측이 수행되고 있으며, 예측기간과 필요 정확도에 따라 초단기, 단기, 중 장기 예측 등으로 구분할 수 있다. 유량예측에 사용되는 기법들은 기후변화 시나리오와 같이 예측된 강우자료를 이용하여 유출량을 예측하는 방법이 있으며, 통계적인 방법으로 과거자료들을 활용하여 미래의 유량을 예측하는 방법이 있다. 본 연구에서는 ESP 기법을 이용하여 금호강 유역의 월 단위(30일) 유량을 예측하고자 한다. 앙상블 유량예측기법(ESP; Ensemble Streamflow Prediction)이란 현재의 유역상태를 초기조건으로 사용하고 과거의 온도나 강수 등의 시계열 앙상블을 강우-유출모형에 입력하여 유출량을 앙상블로 예측하는 기법이다. ESP는 결국 현재의 유역상태와 유역에서의 과거 강우 관측기록, 미래 강우예측에 대한 정보를 조합하여 그에 따른 유출 앙상블을 생산해내게 된다. 월 유량을 예측하기 위해서 금호강 유역의 1988년에서 2014년까지 27년간 대구, 영천, 포항 관측소의 기상자료를 수집하였으며, 금호강 표준유역에 해당하는 19개 유역으로 분할하여 모의에 이용하였다. 금호강 유역에 티센망을 적용하여 각 표준유역별로 강우량을 조합하여 2013년까지 모의에 적용하였으며, 이는 과거자료로 사용하였다. 유량예측에 사용되는 강우자료를 생성하기 위해서 26년간 일강우를 이용하였다. 예를 들어 2014년 12월을 예측한다면 11월까지 관측된 유역초기 조건을 가지는 수문모형의 12월 기상입력자료로써 현재 유역에서 발생 가능성이 있는 동일 유역의 과거 1988년부터 2013년까지의 12월 기상자료들을 사용하는 방법이다. 1988년부터 2013년까지 26개 12월 기상자료를 사용하므로 유량예측결과 또한 26개가 주워진다. 계산된 26개의 유량앙상블이 적용된 유역에서 12월에 발생 가능한 유출량의 모음이 된다. 시나리오결과를 수자원관리에 활용하기 위해서 초과확률로 분석하였으며, 이런 분석의 결과는 향후 가뭄과 홍수 같은 수해방지를 위해 수공구조물의 운영에도 활용할 수 있을 것으로 판단된다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.3
• /
• pp.20-29
• /
• 2021

In the Industry 4.0 era, artificial intelligence has attracted considerable interest for learning mass data to improve the accuracy of forecasting and classification. On the other hand, the current method of detecting anomalies relies on traditional statistical methods for a limited amount of data, making it difficult to detect accurate anomalies. Therefore, this paper proposes an artificial intelligence-based anomaly detection methodology to improve the prediction accuracy and identify new data patterns. In particular, data were collected and analyzed from the point of view that sensor data collected at vehicle idle could be used to detect abnormalities. To this end, a sensor was designed to determine the appropriate time length of the data entered into the forecast model, compare the results of idling data with the overall driving data utilization, and make optimal predictions through a combination of various sensor data. In addition, the predictive accuracy of artificial intelligence techniques was presented by comparing Convolutional Neural Networks (CNN) and Long Short Term Memory (LSTM) as the predictive methodologies. According to the analysis, using idle data, using 1.5 times of the data for the idling periods, and using CNN over LSTM showed better prediction results.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.8
• /
• pp.1005-1012
• /
• 2021

Recently, the use of various location-based services-based location information systems using maps on the web has been expanding, and there is a need for a monitoring system that can check power demand in real time as an alternative to energy saving. In this study, we developed a deep learning real-time virtual power demand prediction web system using open source-based mapping service to analyze and predict the characteristics of power demand data using deep learning. In particular, the proposed system uses the LSTM(Long Short-Term Memory) deep learning model to enable power demand and predictive analysis locally, and provides visualization of analyzed information. Future proposed systems will not only be utilized to identify and analyze the supply and demand and forecast status of energy by region, but also apply to other industrial energies.

• Structural Engineering and Mechanics
• /
• v.81 no.6
• /
• pp.665-675
• /
• 2022

The safety problems of giant hydraulic structures such as dams caused by terrorist attacks, earthquakes, and wars often have an important impact on a country's economy and people's livelihood. For the national defense department, timely and effective assessment of damage to or impending damage to dams and other structures is an important issue related to the safety of people's lives and property. In the field of damage assessment and vulnerability analysis, it is usually necessary to give the damage assessment results within a few minutes to determine the physical damage (crack length, crater size, etc.) and functional damage (decreased power generation capacity, dam stability descent, etc.), so that other defense and security departments can take corresponding measures to control potential other hazards. Although traditional numerical calculation methods can accurately calculate the crack length and crater size under certain combat conditions, it usually takes a long time and is not suitable for rapid damage assessment. In order to solve similar problems, this article combines simulation calculation methods with machine learning technology interdisciplinary. First, the common concrete gravity dam shape was selected as the simulation calculation object, and XFEM (Extended Finite Element Method) was used to simulate and calculate 19 cracks with different initial positions. Then, an LSTM (Long-Short Term Memory) machine learning model was established. 15 crack paths were selected as the training set and others were set for test. At last, the LSTM model was trained by the training set, and the prediction results on the crack path were compared with the test set. The results show that this method can be used to predict the crack propagation path rapidly and accurately. In general, this article explores the application of machine learning related technologies in the field of mechanics. It has broad application prospects in the fields of damage assessment and vulnerability analysis.

• Journal of Aerospace System Engineering
• /
• v.16 no.4
• /
• pp.45-52
• /
• 2022

In the field of UAM, which is attracting attention as a next-generation transportation system, technology developments for using UAVs have been actively conducted in recent years. Since UAVs adopted with these technologies are mainly operated in urban areas, it is imperative that accidents are prevented. However, it is not easy to predict the abnormal flight state of an UAV causing a crash, because of its strong non-linearity. In this paper, we propose a method for predicting a flight state of an UAV, based on a CNN-LSTM hybrid model. To predict flight state variables at a specific point in the future, the proposed model combines the CNN model extracting temporal and spatial features between flight data, with the LSTM model extracting a short and long-term temporal dependence of the extracted features. Simulation results show that the proposed method has better performance than the prediction methods, which are based on the existing artificial neural network model.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.3
• /
• pp.42-52
• /
• 2022

Bus Information System (BIS) services are expanding nationwide to small and medium-sized cities, including large cities, and user satisfaction is continuously improving. In addition, technology development related to improving reliability of bus arrival time and improvement research to minimize errors continue, and above all, the importance of information accuracy is emerging. In this study, accuracy performance was evaluated using LSTM, a machine learning method, and compared with existing methodologies such as Kalman filter and neural network. As a result of analyzing the standard error for the actual travel time and predicted values, it was analyzed that the LSTM machine learning method has about 1% higher accuracy and the standard error is about 10 seconds lower than the existing algorithm. On the other hand, 109 out of 162 sections (67.3%) were analyzed to be excellent, indicating that the LSTM method was not entirely excellent. It is judged that further improved accuracy prediction will be possible when algorithms are fused through section characteristic analysis.

• Korean Journal of Agricultural Science
• /
• v.49 no.2
• /
• pp.193-202
• /
• 2022

The recurrent neural network (RNN) algorithm has been widely used in water-related research areas, such as water level predictions and water quality predictions, due to its excellent time series learning capabilities. However, studies on water quality predictions using RNN algorithms are limited because of the scarcity of water quality data. Therefore, most previous studies related to water quality predictions were based on monthly predictions. In this study, the quality of the water in a reservoir in Nonsan, Chungcheongnam-do Republic of Korea was predicted using the RNN-LSTM algorithm. The study was conducted after constructing data that could then be, linearly interpolated as daily data. In this study, we attempt to predict the water quality on the 7th, 15th, 30th, 45th and 60th days instead of making daily predictions of water quality factors. For daily predictions, linear interpolated daily water quality data and daily weather data (rainfall, average temperature, and average wind speed) were used. The results of predicting water quality concentrations (chemical oxygen demand [COD], dissolved oxygen [DO], suspended solid [SS], total nitrogen [T-N], total phosphorus [TP]) through the LSTM algorithm indicated that the predictive value was high on the 7th and 15th days. In the 30th day predictions, the COD and DO items showed R² that exceeded 0.6 at all points, whereas the SS, T-N, and T-P items showed differences depending on the factor being assessed. In the 45th day predictions, it was found that the accuracy of all water quality predictions except for the DO item was sharply lowered.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.358-358
• /
• 2023

본 연구에서는 그동안 수자원분야 강우유출 해석분야에 활용되었던 대표적인 머신러닝&딥러닝(ML&DL) 모델을 활용하여 모델의 하이퍼파라미터 튜닝뿐만 아니라 모델의 특성을 고려한 기상 및 수문데이터의 조합과 전처리(lag-time, 이동평균 등)를 통하여 데이터 특성과 ML&DL모델의 조합시나리오에 따른 일 유입량 예측성능을 비교 검토하는 연구를 수행하였다. 이를 위해 소양강댐 유역을 대상으로 1974년에서 2021년까지 축적된 기상 및 수문데이터를 활용하여 1) 강우, 2) 유입량, 3) 기상자료를 주요 영향변수(독립변수)로 고려하고, 이에 a) 지체시간(lag-time), b) 이동평균, c) 유입량의 성분분리조건을 적용하여 총 36가지 시나리오 조합을 ML&DL의 입력자료로 활용하였다. ML&DL 모델은 1) Linear Regression(LR), 2) Lasso, 3) Ridge, 4) SVR(Support Vector Regression), 5) Random Forest(RF), 6) LGBM(Light Gradient Boosting Model), 7) XGBoost의 7가지 ML방법과 8) LSTM(Long Short-Term Memory models), 9) TCN(Temporal Convolutional Network), 10) LSTM-TCN의 3가지 DL 방법, 총 10가지 ML&DL모델을 비교 검토하여 일유입량 예측을 위한 가장 적합한 데이터 조합 특성과 ML&DL모델을 성능평가와 함께 제시하였다. 학습된 모형의 유입량 예측 결과를 비교·분석한 결과, 소양강댐 유역에서는 딥러닝 중에서는 TCN모형이 가장 우수한 성능을 보였고(TCN>TCN-LSTM>LSTM), 트리기반 머신러닝중에서는 Random Forest와 LGBM이 우수한 성능을 보였으며(RF, LGBM>XGB), SVR도 LGBM수준의 우수한 성능을 나타내었다. LR, Lasso, Ridge 세가지 Regression모형은 상대적으로 낮은 성능을 보였다. 또한 소양강댐 댐유입량 예측에 대하여 강우, 유입량, 기상계열을 36가지로 조합한 결과, 입력자료에 lag-time이 적용된 강우계열의 조합 분석에서 세가지 Regression모델을 제외한 모든 모형에서 NSE(Nash-Sutcliffe Efficiency) 0.8이상(최대 0.867)의 성능을 보였으며, lag-time이 적용된 강우와 유입량계열을 조합했을 경우 NSE 0.85이상(최대 0.901)의 더 우수한 성능을 보였다.