• Journal of Korea Water Resources Association
• /
• v.47 no.11
• /
• pp.1095-1105
• /
• 2014

A spatial downscaling method using the Support Vector Machine (SVM) Regression for 25 km Tropical Rainfall Measuring Mission (TRMM) Monthly precipitation is proposed. The nonlinear relationship among hydrometeorological variables and precipitation was effectively depicted by the SVM for predicting downscaled grid precipitation. The accuracy of spatially downscaled precipitation was estimated by comparing with rain gauge data from sixty-four stations and found to be improved than the original TRMM data in overall. Especially the positive bias of the original TRMM data was effectively removed after the downscaling procedure. The spatial distributions of 25 km and 1 km grid precipitation were generally similar, while the local spatial trend was better detected by 1 km grid precipitation. The downscaled grid data derived from the proposed method can be applied in hydrological modelling for higher accuracy and further be studied for developing optimized downscaling method incorporation other regression methods.

• The Korean Journal of Applied Statistics
• /
• v.32 no.6
• /
• pp.795-835
• /
• 2019

Big data has been generated in various fields. Many companies have now tried to make profits by building a system capable of analyzing big data based on artificial intelligence (AI) techniques. Integrating AI technology has made analyzing and utilizing vast amounts of data increasingly valuable. In particular, demand forecasting with maximum accuracy is critical to government and business management in various fields such as finance, procurement, production and marketing. In this case, it is important to apply an appropriate model that considers the demand pattern for each field. It is possible to analyze complex patterns of real data that can also be enlarged by a traditional time series model or regression model. However, choosing the right model among the various models is difficult without prior knowledge. Many studies based on AI techniques such as machine learning and deep learning have been proven to overcome these problems. In addition, demand forecasting through the analysis of stereotyped data and unstructured data of images or texts has also shown high accuracy. This paper introduces important areas where demand forecasts are relatively active as well as introduces machine learning and deep learning techniques that consider the characteristics of each field.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.73-73
• /
• 2019

지구 표면의 약 2%에 해당하는 담수에서 육상계 전체가 흡수하는 탄소의 50%가 배출되며, 이는 토양표면에서 배출되는 탄소량에 비해 더 큰 수치로 전 지구적 탄소순환 해석에 중요한 역할을 한다. 특히, 내륙수역과 대기의 경계면에서 $CO_2$ 이동은 전 지구적 탄소순환의 중요한 구성요소로 평가되고 있다. 호수와 저수지 같은 담수 저류시설은 육상에서 기인한 탄소의 운송 및 처리 역할을 한다. 하지만, 저수지에서 온실가스배출량을 평가할 수 있는 명확한 방법론이 부족하며, 전지구 규모 GHGs배출량에 대한 추정에 대한 불확실성이 상당히 큰 상황이다. 본 연구에서는 몬순기후대에 위치한 인공저수지를 대상으로 보다 신뢰도있는 온실가스 배출량 추정을 위해 $CO_2$ NAF 산정하고, 산정에 영향을 미치는 인자들을 분석 하였다. 분석을 위해 $CO_2$ NAF 산정에 필요한 수리 및 수질 인자들을 2017년부터 2018년까지 수집하고, 기초통계량 및 상관분석을 실시하였다. 또한, 주성분분석(PCA) 및 다중선형회귀모델(MLR)과 랜덤포레스트(RF) 기법을 사용해 변수 중요도를 평가하였으며, $CO_2$ NAF 산정 주요인자인 기체교환 계수를 경험적 모델 3종(Cole and Caraco, Crusius, Vachon), 표면갱신형 모델 4종(Heiskanen, Maclntyre, Read, Soloviev)을 비교, 검토하였다. 조사기간 동안 기체교환계수 산정 결과 Crusius 모델 예측값이 평균 $0.342(0.047{\sim}4.323)cm\;hr^{-1}$으로 검토한 모델중 가장 낮은 평균값을 보였으며, Heiskane 모델이 $2.135(0.337{\sim}5.152)cm\;hr^{-1}$으로 가장 큰 평균값을 보였다. 대상 수체는 연주기로 완전혼합되며 수온성층이 약화되는 시기에 저수지 표층 아래에 축적된 탄소가 표층으로 전달되어 높은 수준의 p$CO_2$를 보이며, 수표면에 큰 난류 강도가 작용하는 기간에 대기중으로 배출(pulse emission) 기작이 나타난다. NAF 산정결과 경험적 모델의 NAF값($-1246.0{\sim}6510.3mg-CO_2m^{-2}day^{-1}$)은 표면갱신형 모델 NAF값($-1436.1{\sim}8485.7mg-CO_2m^{-2}day^{-1}$)보다 낮은 수준을 보였으며, 풍속의 함수만을 이용하는 경험적 모델보다 부력 플럭스와 난류 혼합의 영향을 고려하는 Macintyre, Heiskanen모델이 성층 저수지의 $CO_2$ NAF 산정에 적합한 것으로 나타났다. $CO_2$ NAF 산정의 주요인자로 MLR모델은 Tw, EC, pH, Chla, TOC, Alk, RF모델은 EC, DO, TOC가 중요 변수로 평가되었다. PCA 분석결과, 수온이 낮고 성층이 약화되며 pH가 낮은 상태에서 NAF가 큰 것으로 나타났다.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_3
• /
• pp.1711-1720
• /
• 2020

Particulate matter (PM) that has been artificially generated during the recent of rapid industrialization and urbanization moves and disperses according to weather conditions, and adversely affects the human skin and respiratory systems. The purpose of this study is to predict the PM10 concentration in Seoul using meteorological factors as input dataset for multiple linear regression (MLR), support vector machine (SVM), and random forest (RF) models, and compared and evaluated the performance of the models. First, the PM10 concentration data obtained at 39 air quality monitoring sites (AQMS) in Seoul were divided into training and validation dataset (8:2 ratio). The nine meteorological factors (mean, maximum, and minimum temperature, precipitation, average and maximum wind speed, wind direction, yellow dust, and relative humidity), obtained by the automatic weather system (AWS), were composed to input dataset of models. The coefficients of determination (R2) between the observed PM10 concentration and that predicted by the MLR, SVM, and RF models was 0.260, 0.772, and 0.793, respectively, and the RF model best predicted the PM10 concentration. Among the AQMS used for model validation, Gwanak-gu and Gangnam-daero AQMS are relatively close to AWS, and the SVM and RF models were highly accurate according to the model validations. The Jongno-gu AQMS is relatively far from the AWS, but since PM10 concentration for the two adjacent AQMS were used for model training, both models presented high accuracy. By contrast, Yongsan-gu AQMS was relatively far from AQMS and AWS, both models performed poorly.

• Journal of the Korea Institute of Building Construction
• /
• v.13 no.5
• /
• pp.491-497
• /
• 2013

The objective of the present study is to establish a straightforward mixture proportioning procedure for structural lightweight aggregate concrete (LWAC), and evaluate the selection range of the targeted dry density of concrete against the designed concrete compressive strength. In developing this procedure, mathematical models were formulated based on a nonlinear regression analysis over 347 data sets and two boundary conditions of the absolute volume and dry density of concrete. The proposed procedure demonstrated the appropriate water-to-cement ratio and dry density of concrete to achieve the designed strength decrease with the increase in volumetric ratio of coarse aggregates. This trend was more significant in all-LWAC than in sand-LWAC. Overall, the selection range of the dry density of LWAC exists within a certain range according to the designed strength, which can be obtained using the proposed procedure.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.4
• /
• pp.649-654
• /
• 2009

Soil temperature is an important tool in predicting a change of climate and agricultural environment together with the change of atmospheric temperature. In this paper, we examine changing patterns of soil temperature measured in 0.5m under ground from 1932 to 1990 and atmospheric temperature from 1961 to 2008, and derive a relationship between atmospheric temperature and soil temperature. Using this model, we predict unmeasured soil temperature in Daegu area and soil temperature is found to be increasing about $0.028^{\circ}C$per a year. Prediction of soil temperature is an important indicator for climate change in Daegu and will be useful information to help us take precautions for global warming, etc.

• International Journal of Highway Engineering
• /
• v.6 no.4 s.22
• /
• pp.109-121
• /
• 2004

This study was performed to investigate the characteristics of drying shrinkage for concrete slabs as a project for Korean pavement design procedure. According to the volume-surface ratios and aggregate types, the experiments have been executed for 252 days. In order to simulate the volume-surface ratio of a real concrete pavement slab, three-layer epoxy coating and wrapping were used to prevent the evaporation at the part of specimen surfaces. As a result of preliminary test, coating and wrapping method was identified as reliable for three months. According to the volume-surface ratio, the drying shrinkage of the concrete specimen using sandstone was measured 1.32 to 1.8 times higher than that of the limestone specimen. Comparing to the measured drying shrinkage strains and established ACI and CEB-FIP model equations, it turned out that those model equations were underestimated. Finally, considering the age and volume-surface ratios, the prediction equations of the drying shrinkage of concrete specimen were proposed through a multiple nonlinear regression analysis.

• Journal of the Korea Concrete Institute
• /
• v.28 no.4
• /
• pp.427-434
• /
• 2016

The purpose of this study is to propose a mixture proportioning approach based on the replacement level of natural sand for reducing $CO_2$ emissions from artificial lightweight aggregate concrete(LWAC) production. To assess the effect of natural sand on the reduction of $CO_2$ emissions and compressive strength of LWAC, a total of 379 specimens compiled from different sources were analyzed. Based on the non-linear regression analysis using the database and the previous mixture proportioning method proposed by Yang et al., simple equations were derived to determine the concrete mixture proportioning and the replacement level of natural sand for achieving the targeted performances(compressive strength, initial slump, air content, and $CO_2$ reduction ratio) of concrete. Furthermore, the proposed equations are practically applicable to straightforward determination of the $CO_2$ emissions from the provided mixture proportions of LWAC.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.31 no.1C
• /
• pp.11-17
• /
• 2011

This study was performed a series of the isotropic compression-expansion tests and the drained triaxial tests with various the relative densities 25%, 50%, 80% and 100% for Baekma river sand. Using the tests results the characteristic of the parameters of Lade's single hardening constitutive model were investigated. The soil parameters Kur and n representing elastic behavior are not much affected by the change of the relative density. The other parameters such as failure criterion (m, ${\eta}_1$), hardening function (C, p) and plastic potential (${\Psi}_2$, ${\mu}$) are in a positive linear relationship with the relative density. Since the soil parameters h and $\alpha$ representing yield function do not change much to the change of relative density and also closely related to failure criterion, they can be replaced by failure criterion ${\eta}_1$. We also observed that predicted values from the Lade's single hardening constitutive model were well consistent with the observed data.

• Smart Media Journal
• /
• v.11 no.4
• /
• pp.19-29
• /
• 2022

Recently, with the development of IoT technology, the number of farms using smart farms is increasing. Smart farms monitor the environment and optimise internal environment automatically to improve crop yield and quality. For optimized crop cultivation, researches on predict crop productivity are actively studied, by using collected agricultural digital data. However, most of the existing studies are based on statistical models based on existing statistical data, and thus there is a problem with low prediction accuracy. In this paper, we use various predition models for predicting the production and sales profits, and compare the performance results through models by using the agricultural digital data collected in the facility horticultural smart farm. The models that compared the performance are multiple linear regression, support vector machine, artificial neural network, recurrent neural network, LSTM, and ConvLSTM. As a result of performance comparison, ConvLSTM showed the best performance in R2 value and RMSE value.