• Journal of Astronomy and Space Sciences
• /
• v.27 no.4
• /
• pp.367-375
• /
• 2010

In this study, we compared the precipitable water vapor (PWV) data derived from the radiosonde observation data at Sokcho Observatory and the PWV data at Sokcho Global Positioning System (GPS) Observatory provided by Korea Astronomy and Space Science Institute, from 0000 UTC, June 1, 2007 to 1200 UTC, May 31, 2009, and analyzed the radiosonde bias between the day and the night. In the scatter diagram of the daytime and nighttime radiosonde PWV data and the GPS PWV data, dry bias was found in the daytime radiosonde observation as known in the previous study. In addition, for all the rainfall events, the tendency that the wet bias of the radiosonde PWV increased as the GPS PWV decreased and the dry bias of the radiosonde PWV increased as the GPS PWV increased was significantly less distinctive in nighttime than in daytime. The quantitative analysis of the bias and error of the radiosonde PWV data showed that the mean bias decreased in the second year, regardless of nighttime or daytime rainfall, and the non-rainfall root mean square error (RMSE) was similar to that of the previous studies, while the rainfall RMSE was larger to a certain extent.

• KIEAE Journal
• /
• v.11 no.1
• /
• pp.15-20
• /
• 2011

Computer simulation of buildings and solar energy systems is being used increasingly in energy assessments and design. For the six locations (Seoul, Incheon, Daejeon, Deagu, Gwangju and Busan) in South Korea where the global hourly solar irradiation (GHSI) is currently measured, GHSI was calculated using a comparatively simple cloud cover radiation model (CRM) and sunshine fraction radiation model (SFRM). The result was that the measured and calculated values of GHSI were similar for the six regions. Results of cloud cover and sunshine fraction models have been compared with the measured data using the coefficient of determination (R2), root-mean-square error (RMSE) and mean bias error (MBE). The strength of correlation R2 varied within similar ranges: 0.886-0.914 for CRM and 0.908-0.934 for SFRM. Average MBE for the CRM and SFRM were 6.67 and 14.02 W/m2, respectively, and average RMSE 104.36 and 92.15 W/m2. This showed that SFRM was slightly accurate and used many regions as compared to CRM for prediction of GHSI.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.51 no.1
• /
• pp.1-9
• /
• 2009

The objective of this paper was to evaluate the auto-calibration with multi-objective optimization method to calibrate the parameters of the Soil and Water Assessment Tool (SWAT) model. The model was calibrated and validated by using nine years (1996-2004) of measured data for the 384-ha Baran reservoir subwatershed located in central Korea. Multi-objective optimization was performed for sixteen parameters related to runoff. The parameters were modified by the replacement or addition of an absolute change. The root mean square error (RMSE), relative mean absolute error (RMAE), Nash-Sutcliffe efficiency index (EI), determination coefficient ($R^2$) were used to evaluate the results of calibration and validation. The statistics of RMSE, RMAE, EI, and $R^2$ were 4.66 mm/day, 0.53 mm/day 0.86, and 0.89 for the calibration period and 3.98 mm/day, 0.51 mm/day, 0.83, and 0.84 for the validation period respectively. The statistical parameters indicated that the model provided a reasonable estimation of the runoff at the study watershed. This result was illustrated with a multi-objective optimization for the flow at an observation site within the Baran reservoir watershed.

• Journal of Forest and Environmental Science
• /
• v.39 no.4
• /
• pp.195-202
• /
• 2023

In this study, we sought to compare and evaluate the accuracy and predictive performance of machine learning algorithms for estimating the growth of individual Larix kaempferi trees in Gangwon Province, Korea. We employed linear regression, random forest, XGBoost, and LightGBM algorithms to predict tree growth using monitoring data organized based on different thinning intensities. Furthermore, we compared and evaluated the goodness-of-fit of these models using metrics such as the coefficient of determination (R²), mean absolute error (MAE), and root mean square error (RMSE). The results revealed that XGBoost provided the highest goodness-of-fit, with an R² value of 0.62 across all thinning intensities, while also yielding the lowest values for MAE and RMSE, thereby indicating the best model fit. When predicting the growth volume of individual trees after 3 years using the XGBoost model, the agreement was exceptionally high, reaching approximately 97% for all stand sites in accordance with the different thinning intensities. Notably, in non-thinned plots, the predicted volumes were approximately 2.1 m³ lower than the actual volumes; however, the agreement remained highly accurate at approximately 99.5%. These findings will contribute to the development of growth prediction models for individual trees using machine learning algorithms.

• Computers and Concrete
• /
• v.34 no.4
• /
• pp.477-487
• /
• 2024

The in-situ stresses of concrete are an essential index for assessing the safety performance of concrete structures. Conventional methods for pore pressure release often face challenges in selecting drilling ring parameters, uncontrollable stress release, and unstable detection accuracy. In this paper, the parameters affecting the results of the concrete ring hole stress release method are cross-combined, and finite elements are used to simulate the combined parameters and extract the stress release values to establish a training set. The GridSearchCV function is utilized to determine the optimal hyperparameters. The mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R²) are used as evaluation indexes to train the gradient boosting decision tree (GBDT) algorithm, and the other three common algorithms are compared. The RMSE of the GBDT algorithm for the test set is 4.499, and the R² of the GBDT algorithm for the test set is 0.962, which is 9.66% higher than the R² of the best-performing comparison algorithm. The model generated by the GBDT algorithm can accurately calculate the concrete in-situ stresses based on the drilling ring parameters and the corresponding stress release values and has a high accuracy and generalization ability.

• Advances in concrete construction
• /
• v.18 no.2
• /
• pp.135-146
• /
• 2024

All evidence-based waste management endeavour needs accurate data on construction waste creation, but because many developing nations have outdated recording systems, this data is still hard to come by. Around 50% of global carbon dioxide (CO₂) emissions connected to energy use in buildings have historically come from this industry. Thus, in the global endeavour to decarbonise the energy system, it garners a great deal of attention. In order to anticipate CO₂ emissions from buildings over the long term, this research introduces and compares several Machine Learning (ML)-based methods. This research proposes novel technique in concrete waste reduction based on energy efficiency analysis and carbon footprint modelling using machine learning algorithms. Here the concrete construction waste reduction with energy efficiency is carried out using Bayesian multilayer reinforcement neural networks. then the carbon footprint analysis in smart building construction using fuzzy Gaussian linear hidden markov vector model. the experimental analysis has been carried out based on various concrete composition and CO₂ analysis in terms of MAPE (mean average energy efficiency error), detection accuracy, correlation coefficient values (R), root mean square error (RMSE), energy efficiency. Proposed method produced 98% detection accuracy, 97% correlation coefficient values, 95% energy efficiency, 68% RMSE, and 58% MAPE.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.33 no.3
• /
• pp.145-153
• /
• 2015

This study aims to accurately estimate population distribution more specifically than administrative unites using a RK (Regression-Kriging) model. The RK model is the areal interpolation technique that involves linear regression and the Kriging model. In order to estimate a population’s distribution using a sample region, four different models were used, namely; a regression model, RK model, OK (Ordinary Kriging) model and CK (Co-Kriging) model. The results were then compared with each other. Evaluation of the accuracy and validity of evaluation analysis results were the basis RMSE (Root Mean Square Error), MAE (Mean Absolute Error), G statistic and correlation coefficient (ρ). In the sample regions, every statistic value of the RK model showed better results than other models. The results of this comparative study will be useful to estimate a population distribution of the metropolitan areas with high population density

• Journal of the Korean Solar Energy Society
• /
• v.30 no.2
• /
• pp.16-21
• /
• 2010

Computer simulation of buildings and solar energy systems are being used increasingly in energy assessments and design. This paper discusses the possibility of using sunshine duration data instead of global hourly solar irradiation (GHSI) data for localities with abundant data on sunshine duration. For six locations in South Korea where global radiation is currently measured, the global radiation was calculated using Sunshine Duration Radiation Model (SDRM), compared and analyzed. Results of SDRM has been compared with the measured data on the coefficients of determination (R2), root-mean-square error (RMSE) and mean bias error (MBE). This study recommends the use of sunshine duration based irradiation models if measured solar radiation data is not available.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.68-68
• /
• 2022

코로나바이러스는(COVID-19)는 2019년 12일 중국 후베이성 우한시에서 시작된 코로나바이러스감염증으로 2020년 1월부터 전 세계로 퍼져, 일부 국가 및 지역을 제외한 대부분의 나라와 모든 대륙으로 확산되었다. 이에 WHO는 범 유행전염병(Pandemic)을 선언하였다. 2022년 3월 18일 현재 국내 누적 확진환자 8,657,609명과 11,782명의 사망자를 일으켰고 전 세계적으로도 많은 사상자를 내고 있는 실정이고 사회 및 경제적인 피해로도 계속 확대되고 있다. 많은 감염자와 사망자의수에 대한 예측은 코로나바이러스의 전염병을 예방하고 즉각적 조치를 취할 수 있는데 도움이 될 수 있다. 본 연구에서는 문화적 인자를 제외한 국내에서 연구 사례가 많지 않은 기상 요인을 인자로 포함하여 머신러닝 모델을 통해 확진자를 예측하였다. 그리고 여러 가지 모델을 성능 평가 기법인 Root Mean Square Error(RMSE) 및 Mean Absolute Percentage Error(MAPE)를 통해 성능을 평가하고 비교하여 정확도 높은 모델을 제시하였다.