• 한국멀티미디어학회논문지
• /
• 제20권8호
• /
• pp.1406-1420
• /
• 2017

Artificial neural networks (ANNs) play an important role in the fields of function approximation, prediction, and classification. ANN performance is critically dependent on the input parameters, including the number of neurons in each layer, and the optimal values of weights and biases assigned to each neuron. In this study, we apply the particle swarm optimization method, a popular optimization algorithm for determining the optimal values of weights and biases for every neuron in different layers of the ANN. Several regression models, including general linear regression, Fourier regression, smoothing spline, and polynomial regression, are conducted to evaluate the proposed method's prediction power compared to multiple linear regression (MLR) methods. In addition, residual analysis is conducted to evaluate the optimized ANN accuracy for both training and test datasets. The experimental results demonstrate that the proposed method can effectively determine optimal values for neuron weights and biases, and high accuracy results are obtained for prediction applications. Evaluations of the proposed method reveal that it can be used for prediction and estimation purposes, with a high accuracy ratio, and the designed model provides a reliable technique for optimization. The simulation results show that the optimized ANN exhibits superior performance to MLR for prediction purposes.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2008년도 추계학술대회A
• /
• pp.1588-1593
• /
• 2008

This paper describes a real-time isometric pinch force prediction algorithm from surface electromyogram (sEMG) using multilayer perceptron (MLP) for human robot interactive applications. The activities of seven muscles which are observable from surface electrodes and also related to the movements of the thumb and index finger joints were recorded during pinch force experiments. For the successful implementation of the real-time prediction algorithm, an off-line analysis was performed using the recorded activities. Four muscles were selected for the force prediction by using the Fisher linear discriminant analysis among seven muscles, and the four muscle activities provided effective information for mapping sEMG to the pinch force. The MLP structure was designed to make training efficient and to avoid both under- and over-fitting problems. The pinch force prediction algorithm was tested on five volunteers and the results were evaluated using two criteria: normalized root mean squared error (NRMSE) and correlation (CORR). The training time for the subjects was only 2 min 29 sec, but the prediction results were successful with NRMSE = 0.112 ${\pm}$ 0.082 and CORR = 0.932 ${\pm}$ 0.058. These results imply that the proposed algorithm is useful to measure the produced pinch force without force sensors in real-time. The possible applications include controlling bionic finger robot systems to overcome finger paralysis or amputation.

• 한국전자파학회논문지
• /
• 제9권2호
• /
• pp.211-218
• /
• 1998

본 논문에서는 광선 추적법을 이용한 마이크로셀 전파 환경 예측 모텔에서의 여러 가지 오차들에 관하여 해석 하였다. 반사 및 회절 계수의 부정확성 및 계산 과정에서 이들의 횟수 제한이 예측 결과에 미치는 영향올 분석하 였으며, 특히 건물 지도의 유한한 해상도에 의한 건물 데이터베이스의 오차가 전파 환경 예측 결과에 미칠 수 있는 영향에 대하여 자세히 분석하였다. 빠르고 정확한 예측 결과를 얻기 위해서는 각 건물들에 대하여 적절한 전기적 상수들의 선정과 반사 및 회절 횟수의 적절한 제한이 필요함을 보였다. 또한, 건물 지도의 해상도가 나빠질 수록 예측 결과의 오차가 증가함을 보였으며, LOS 구간과 NLOS 구간을 구분하여 그 오차의 크기를 분석하였다. 본 논문에서의 결과를 바탕으로 건물 지도의 해상도에 따른 예측 결과의 최대 오차 한계를 6 dB이하로 제한 할 경우 적절한 건물 지도의 해상도는 5m정도가 됨을 알 수 있다.

• 한국가시화정보학회지
• /
• 제21권1호
• /
• pp.18-25
• /
• 2023

In these days, high dimensional data prediction technology based on neural network shows compelling results in many different kind of field including engineering. Especially, a lot of variants of convolution neural network are widely utilized to develop pixel level prediction model for high dimensional data such as picture, or physical field value from the sensors. In this study, velocity vector field of ideal flow on ship surface is estimated on pixel level by Unet. First, potential flow analysis was conducted for the set of hull form data which are generated by hull form transformation method. Thereafter, four different neural network with a U-shape structure were conFig.d to train velocity vectors at the node position of pre-processed hull form data. As a result, for the test hull forms, it was confirmed that the network with short skip-connection gives the most accurate prediction results of streamlines and velocity magnitude. And the results also have a good agreement with potential flow analysis results. However, in some cases which don't have nothing in common with training data in terms of speed or shape, the network has relatively high error at the region of large curvature.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2021년도 학술발표회
• /
• pp.183-183
• /
• 2021

Deep learning methods and their application have become an essential part of prediction and modeling in water-related research areas, including hydrological processes, climate change, etc. It is known that application of deep learning leads to high availability of data sources in hydrology, which shows its usefulness in analysis of precipitation, runoff, groundwater level, evapotranspiration, and so on. However, there is still a limitation on microclimate analysis and prediction with deep learning methods because of deficiency of gauge-based data and shortcomings of existing technologies. In this study, a real-time rainfall prediction model was developed from a sky image data set with convolutional neural networks (CNNs). These daily image data were collected at Chung-Ang University and Korea University. For high accuracy of the proposed model, it considers data classification, image processing, ratio adjustment of no-rain data. Rainfall prediction data were compared with minutely rainfall data at rain gauge stations close to image sensors. It indicates that the proposed model could offer an interpolation of current rainfall observation system and have large potential to fill an observation gap. Information from small-scaled areas leads to advance in accurate weather forecasting and hydrological modeling at a micro scale.

• 풍력에너지저널
• /
• 제15권2호
• /
• pp.10-22
• /
• 2024

This research is a comprehensive analysis of wind power prediction sensitivity using a Long Short-Term Memory (LSTM) deep learning neural network model, accounting for the inherent uncertainties in wind speed estimation. Utilizing a year's worth of operational data from an operational wind farm, the study forecasts the power output of both individual wind turbines and the farm collectively. Predictions were made daily at intervals of 10 minutes and 1 hour over a span of three months. The model's forecast accuracy was evaluated by comparing the root mean square error (RMSE), normalized RMSE (NRMSE), and correlation coefficients with actual power output data. Moreover, the research investigated how inaccuracies in wind speed inputs affect the power prediction sensitivity of the model. By simulating wind speed errors within a normal distribution range of 1% to 15%, the study analyzed their influence on the accuracy of power predictions. This investigation provided insights into the required wind speed prediction error rate to achieve an 8% power prediction error threshold, meeting the incentive standards for forecasting systems in renewable energy generation.

• 지능정보연구
• /
• 제22권2호
• /
• pp.33-56
• /
• 2016

대부분의 부도 예측에 관한 연구는 재무 변수를 중심으로 통계적 방법 또는 인공지능 기법을 적용하여 부도 예측 모형을 구축하였다. 그러나 재무비율과 같은 회계 정보를 이용한 부도 예측 모형은 재무 제표 결산 시점과 신용평가 시점 간 시차를 고려하지 않을 뿐만 아니라 해당 산업의 경제적 상황과 같은 외부 환경적인 요소를 반영하기 어렵다는 한계점이 존재하였다. 기업의 부도 여부를 예측하기 위해 정량 정보인 재무 변수만을 이용하는 것에 한계가 있음에도 불구하고 정성 정보를 부도 예측 모형에 반영한 연구는 아직 미흡한 실정이다. 본 연구에서는 재무 변수를 이용하는 기존 부도 예측 모형의 성과를 개선하기 위해 빅데이터 기반의 정성 정보를 추가적인 입력 변수로 활용하는 부도 예측 모형을 제안하였다. 제안 모형의 성과 향상은 정성 정보를 예측 모형에 통합시키기에 적합한 형태로 정보의 유형을 변환시킬 수 있는가에 따라 달려있다. 이에 본 연구에서는 정성 정보 처리를 위한 방법으로 빅데이터 분석 기법 중 하나인 텍스트 마이닝(Text Mining)을 활용하였다. 해당 산업과 관련된 경제 뉴스 데이터로부터 경제 상황에 대한 감성 정보를 추출하기 위해 도메인 중심의 감성 어휘 사전을 구축하고, 구축된 어휘 사전을 기반으로 감성 분석(Sentiment Analysis)을 수행하였다. 형태소 분석 등을 포함한 텍스트 전처리 과정을 거쳐 감성 어휘를 추출하고, 각 어휘에 대한 극성 및 감성 점수를 부여하였다. 분석 결과, 전통적 부도 예측 모형에 경제 뉴스 데이터에서 도출한 정성 정보를 반영하는 것은 모형의 성과를 개선하는 것으로 나타났다. 특히, 경제 상황에 대한 부정적 감정이 기업의 부도 여부를 예측하는 데 더욱 효과적임을 알 수 있었다.

• 한국통신학회논문지
• /
• 제38C권3호
• /
• pp.263-270
• /
• 2013

최근 세계적으로 바다가 자원의 보고로 주목 받으면서 해양 환경 분석 및 예측 기술에 대한 연구가 활발히 진행 되고 있다. 자동화된 해양 환경 자료의 수집과 수집된 자료를 분석하여서 해양재해를 예측하면 기름 유출에 의한 해양오염의 피해, 적조에 의한 수산업의 피해, 해양환경 이변에 의한 수산업 및 재해 피해를 최소화하는데 기여할 수 있다. 그러나 국내 해양 환경에 대한 조사 및 분석 연구는 제한적이다. 본 논문은 국내의 원해 및 근 해역에서 수집된 해양 환경 자료를 분석하여 해양재해를 예측할 수 있는 시스템 모델을 연구한다. 이를 위해서 본 논문에서는 해양재해 예측 시스템을 위해서 통신시스템 모델, 해양환경 자료 수집 시스템 모델, 예측분석 시스템 모델, 상황전파시스템에 대한 모델을 제시하였다. 또한 예측분석 시스템을 위한 적조 예측 모델과 요약분석 모델을 제시하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권5호
• /
• pp.1963-1978
• /
• 2015

With the advancement of mobile web environments, identification and analysis of the user behavior play a significant role and remains a challenging task to implement with variations observed in the model. This paper presents an efficient method for mining optimized user behavior prediction model using genetic algorithm on mobile web structure. The framework of optimized user behavior prediction model integrates the temporary and permanent register information and is stored immediately in the form of integrated logs which have higher precision and minimize the time for determining user behavior. Then by applying the temporal characteristics, suitable time interval table is obtained by segmenting the logs. The suitable time interval table that split the huge data logs is obtained using genetic algorithm. Existing cluster based temporal mobile sequential arrangement provide efficiency without bringing down the accuracy but compromise precision during the prediction of user behavior. To efficiently discover the mobile users' behavior, prediction model is associated with region and requested services, a method called optimized user behavior Prediction Model using Genetic Algorithm (PM-GA) on mobile web structure is introduced. This paper also provides a technique called MAA during the increase in the number of models related to the region and requested services are observed. Based on our analysis, we content that PM-GA provides improved performance in terms of precision, number of mobile models generated, execution time and increasing the prediction accuracy. Experiments are conducted with different parameter on real dataset in mobile web environment. Analytical and empirical result offers an efficient and effective mining and prediction of user behavior prediction model on mobile web structure.

• 한국콘크리트학회:학술대회논문집
• /
• 한국콘크리트학회 1999년도 학회창립 10주년 기념 1999년도 가을 학술발표회 논문집
• /
• pp.285-288
• /
• 1999

Creep is a major parameter to represent long-term behavior of concrete structures concerning serviceability and durability. The effect of creep is recently taking account into crack resistance analysis of early-age concrete concerning durability evaluation. Since existing creep prediction models were proposed to predict creep for hardened concrete, most of them cannot consider effectively the information on microstructure formation and hydration developed in the early-age concrete. In this study, creep tests for early-age concrete made of the type I cement and the type V cement are carried out respectively and creep prediction models are evaluated for the prediction of creep behavior in early-age concrete. A creep prediction model is modified for the prediction of creep in early-age concrete and also verified by comparing prediction results with results of creep tests on early-age concrete.