• 한국인터넷방송통신학회논문지
• /
• 제19권5호
• /
• pp.79-85
• /
• 2019

본 연구에서는 Long Short Term Memory (LSTM) 신경망과 Gated Recurrent Unit(GRU) 신경망을 Internet of Things (IoT) 파워미터에 적용하여 단기 전력사용량 예측방법을 제안하고, 실제 가정의 전력사용량 데이터를 토대로 예측 성능을 분석한다. 성능평가 지표로써 Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Mean Percentage Error (MPE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE)를 이용한다. 실험 결과는 GRU 기반의 모델이 LSTM 기반의 모델에 비해 MAPE 기준으로 4.52%, MPE 기준으로 5.59%만큼의 성능개선을 보였다.

• 한국산학기술학회논문지
• /
• 제11권8호
• /
• pp.2793-2800
• /
• 2010

이 논문에서는 여름철 일일 전력수요 총량을 예측하는 회귀모형을 개발한다. 경제적인 전력 생산계획을 수립하기위해 예측 오차율을 낮추는 것은 매우 중요하다. 전력수요가 크게 증가하는 여름철 전력수요를 예측하기위해 기존 연구에서는 외기온도 및 직전일 전력수요를 고려하였으나, 이 논문에서는 기존 연구에서 제시한 예측 오차율을 개선하기 위해 전력수요의 요일별 특성을 추가적으로 고려한 회귀모형을 개발한다. 이 논문에서는 여름철 전력수요의 요일별 패턴은 최고차항의 계수가 음수인 2차 함수 형태를 나타냄을 확인하였다. 즉, 2005년부터 2009년까지 5년간의 여름철 전력수요 패턴을 살펴본 결과 전력수요 총량은 일요일에 가장 낮고 월요일부터 증가하다가 수요일이나 목요일부터 다시 감소하는 패턴을 보인다. 이 논문에서 제안하는 여름철 전력수요 예측 회귀모형의 타당성을 검증하기 위해 2005년부터 2009년까지 실제 전력수요 데이터를 바탕으로 여름철 전력수요 총량을 예측한 결과, 평균 오차율(MAPE: Mean Absolute Percentage Error)과 최대 오차율(MPE: Maximum Percentage Error)이 각각 3.08%와 8.99%를 넘지 않는 수준임을 확인하였다. 또한 기존 연구에서 제시한 방법과 비교하여도 평균 오차율과 최대 오차율 모두 기존 연구에서 제시한 오차율보다 우수함을 확인하였다.

• 한국도로학회논문집
• /
• 제15권6호
• /
• pp.79-91
• /
• 2013

PURPOSES : The methods of measuring the sound from the noise source are Pass-by method and NCPX (Noble Close Proximity) method. These measuring methods were used to determine the linkage of TAPL (Total Acoustic Pressure Level) and SPL (Sound Pressure Level) in terms of frequencies. METHODS : The frequency analysis methods are DFT (Discrete Fourier Transform) and FFT (Fast Fourier Transform), CPB (Constant Percentage Bandwidth). The CPB analysis was used in this study, based on the 1/3 octave band option configured for the frequency analysis. Furthermore, the regression analysis was used at the condition related to the sound attenuation effect. The MPE (Mean Percentage Error) and RMSE (Root Mean Squared Error) were utilized for calculating the error. RESULTS : From the results of the CPB frequency analysis, the predicted SPL along the frequency has 99.1% maximum precision with the measured SPL, resulting in roughly 1 dB(A) error. The TAPL results have precision by 99.37% with the measured TAPL. The predicted TAPL results at this study by using the SPL prediction model along the frequency have the maximum precision of 98.37% with the vehicle velocity. CONCLUSIONS : The Predicted SPL model along the frequency and the TAPL result by using the predicted SPL model have a high level of accuracy through this study. But the vehicle velocity-TAPL prediction model from the previous study by using the log regression analysis cannot be consistent with the TAPL result by using the predicted SPL model.

• 농업과학연구
• /
• 제38권1호
• /
• pp.129-143
• /
• 2011

In this study, to select the incoming solar radiation equation which is most suitable for the estimation of Penman evaporation, 12 incoming solar radiation equations were selected. The Penman evaporation rates were estimated using 12 selected incoming solar radiation equations, and the estimated Penman evaporation rates were compared with measured pan evaporation rates. The monthly average daily meteorological data measured from 17 meteorological stations (춘천, 강능, 서울, 인천, 수원, 서산, 청주, 대전, 추풍령, 포항, 대구, 전주, 광주, 부산, 목포, 제주, 진주) were used for this study. To evaluate the reliability of estimated evaporation rates, mean absolute bias error(MABE), root mean square error(RMSE), mean percentage error(MPE) and Nash-Sutcliffe equation were applied. The study results indicate that to estimate pan evaporation using Penman evaporation equation, incoming solar radiation equation using meteorological data such as precipitation, minimum air temperature, sunshine duration, possible duration of sunshine, and extraterrestrial radiation are most suitable for 11 study stations out of 17 study stations.

• 한국정보통신학회논문지
• /
• 제25권7호
• /
• pp.884-889
• /
• 2021

국내 코로나19의 감염자 수가 백신과 사회적 거리 두기, 백신 등 여러 가지 노력 덕분에 차츰 줄어드는 듯 보였으나 2020년 2월 20일 특정한 사건 이후 감염자 수가 증가한 것처럼, 2020년 12월부터 또다시 급격히 감염자 수가 증가하는 추세이며 꾸준히 일일 500명가량의 감염자 수가 이어지고 있다. 따라서 Kaggle의 데이터셋을 이용해서 Prophet 알고리즘을 통해 미래 코로나19를 예측하고 사이킷런을 통해 결정계수, 평균 절대 오차, 평균 백분율 오차, 평균 제곱 차, 평균 제곱근 편차를 통해 이 예측에 대한 설명력을 더한다. 또한 코로나19가 급격히 특정한 사건이 없었을 경우 국내 감염자 수를 예측해 앞으로 우리가 미래의 질병에 대해서 방역과 방역 수칙 실천의 중요함을 강조한다.

• 한국도로학회논문집
• /
• 제15권3호
• /
• pp.53-63
• /
• 2013

PURPOSES : This study is to evaluate moisture susceptibility of asphalt mixtures by using non-destructive impact wave and to determine durability so as to decrease the gap between before and after freezing in the future. METHODS : Using non-destructive impact wave, this study is to determine the dynamic modulus of asphalt specimen. Furthermore, the results obtained from two experiment accelerometers are used for the dynamic modulus determination. The dynamic moduli of specimens are compared with those of the freezing-thawing specimens. RESULTS : Test results showed that the dynamic modulus before freezing and thawing environment loads at each temperature dropped about 3.7% after the environmental loads. Furthermore, correlation analysis indicates that transition of dynamic modulus at each point is about 89.59%. CONCLUSIONS: Evaluation of asphalt mixtures using non-destructive impact wave has excellent repeatability and simple equipment for the test. Consequently, the method in the study will be useful for evaluating the characteristics of a various asphalt mixtures.

• 한국도로학회논문집
• /
• 제18권6호
• /
• pp.123-130
• /
• 2016

OBJECTIVES : The objective of this research is to determine the moisture resistance of the freeze-thaw process occurring in low-noise porous pavement using either hydrated-lime or anti-freezing agent. Various additives were applied to low-noise porous asphalt, which is actively paved in South Korea, to overcome its disadvantages. Moreover, the optimum contents of hydrated-lime and anti-freezing agent and behavior properties of low-noise porous asphalt layer are determined using dynamic moduli via the freeze-thaw test. METHODS : The low-noise porous asphalt mixtures were made using gyratory compacters to investigate its properties with either hydrated-lime or anti-freezing agent. To determine the dynamic moduli of each mixture, impact resonance test was conducted. The applied standard for the freeze-thaw test of asphalt mixture is ASTM D 6857. The freeze-thaw and impact resonance tests were performed twice at each stage. The behavior properties were defined using finite element method, which was performed using the dynamic modulus data obtained from the freeze-thaw test and resonance frequencies obtained from non-destructive impact test. RESULTS : The results show that the coherence and strength of the low-noise porous asphalt mixture decreased continuously with the increase in the temperature of the mixture. The dynamic modulus of the normal low-noise porous asphalt mixture dramatically decreased after one cycle of freezing and thawing stages, which is more than that of other mixtures containing additives. The damage rate was higher when the freeze-thaw test was repeated. CONCLUSIONS : From the root mean squared error (RMSE) and mean percentage error (MPE) analyses, the addition rates of 1.5% hydrated-lime and 0.5% anti-freezing agent resulted in the strongest mixture having the highest moisture resistance compared to other specimens with each additive in 1 cycle freeze-thaw test. Moreover, the freeze-thaw resistance significantly improved when a hydrated-lime content of 0.5% was applied for the two cycles of the freeze-thaw test. Hence, the optimum contents of both hydrated-lime and anti-freezing agent are 0.5%.