• Journal of Korea Water Resources Association
• /
• v.53 no.11
• /
• pp.929-938
• /
• 2020

In this study, we estimated missing evapotranspiration (ET) data at a eddy-covariance flux tower in the Cheongmicheon farmland site using the Artificial Neural Network (ANN). The ANN showed excellent performance in numerical analysis and is expanding in various fields. To evaluate the performance the ANN-based gap-filling, ET was calculated using the existing gap-filling methods of Mean Diagnostic Variation (MDV) and Food and Aggregation Organization Penman-Monteith (FAO-PM). Then ET was evaluated by time series method and statistical analysis (coefficient of determination, index of agreement (IOA), root mean squared error (RMSE) and mean absolute error (MAE). For the validation of each gap-filling model, we used 30 minutes of data in 2015. Of the 121 missing values, the ANN method showed the best performance by supplementing 70, 53 and 84 missing values, respectively, in the order of MDV, FAO-PM, and ANN methods. Analysis of the coefficient of determination (MDV, FAO-PM, and ANN methods followed by 0.673, 0.784, and 0.841, respectively.) and the IOA (The MDV, FAO-PM, and ANN methods followed by 0.899, 0.890, and 0.951 respectively.) indicated that, all three methods were highly correlated and considered to be fully utilized, and among them, ANN models showed the highest performance and suitability. Based on this study, it could be used more appropriately in the study of gap-filling method of flux tower data using machine learning method.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.4
• /
• pp.591-598
• /
• 2021

The digital phase-locked loops(DPLL) is a circuit used for phase synchronization and has been generally used in various fields such as communication and circuit fields. State estimators are used to design digital phase-locked loops, and infinite impulse response state estimators such as the well-known Kalman filter have been used. In general, the performance of the infinite impulse response state estimator-based digital phase-locked loop is excellent, but a sudden performance degradation may occur in unexpected situations such as inaccuracy of initial value, model error, and disturbance. In this paper, we propose a minimum variance finite impulse response filter with optimal horizon for designing a new digital phase-locked loop. A numerical method is introduced to obtain the measured value interval length, which is an important parameter of the proposed finite impulse response filter, and to obtain a gain, the covariance matrix of the error is set as a cost function, and a linear matrix inequality is used to minimize it. In order to verify the superiority and robustness of the proposed digital phase-locked loop, a simulation was performed for comparison and analysis with the existing method in a situation where noise information was inaccurate.

• Journal of the Korean School Mathematics Society
• /
• v.24 no.1
• /
• pp.83-105
• /
• 2021

In this paper, we analyzed the misconceptions and errors incurred during factorization learning. We also examined whether online individualization classes had a positive effect on students' mathematical achievement. The experiment was conducted for 4 weeks (16 times in total) on middle school juniors in rural areas of Gyeonggi Province, where the influence of private extra education was small. In the class, the 'Google Classroom' was used as a LMS, the video lecture was uploaded to YouTube, and the teacher interacted with the students through "Zoom" and "Facetalk". In the online class situation, students' assignments and test answers were checked in real time through 'Google Classroom', and immediate feedback was provided to the experimental class group's students. However, for the control group students, feedback was provided only to those who desired. A total of 7 achievement evaluations were conducted in the order of pre-test, formative evaluation (5 times), and post-test to confirm the change in students' ability improvement and achievement. Through the formative evaluation analysis, it was possible to grasp the types of errors and misconceptions that occured during the factorization process. Students' errors were divided into four types: theorem or definition distortion error, functional errors such as calculation, operation, and manipulation, errors that do not verify the solution, and no response. As a result of ANCOVA, the two groups did not show any difference from the 1st to 4th formative assessment. However, the 5th formative assessment and post-test showed statistically significant differences, confirming that online individualization classes contributed to improvemed achievement.

• Journal of the Korean Geographical Society
• /
• v.49 no.5
• /
• pp.761-778
• /
• 2014

There has been relatively little study to model price for commercial property because of its low transaction volume in the market. Despite of this thin market character, this paper tried to estimate prices for commercial lots as accurate as possible. We constructed a model whose components consist of mean structure(global trend), exponential covariance function and a pure error term, and applied it to actual sales price data of Seoul. We explicitly took account of spatial autocorrelation of land price by utilizing a kriging technique, a representative method of spatial interpolation, because the land price of commercial lots has feature of differential price forming pattern depending on submarkets they belong to. In addition, we chose to apply a bayesian kriging to overcome data scarcity by incorporating experts' knowledge into prior probability distribution. The chosen model's excellent performance was verified by the result from validation data. We confirmed that the excellence of the model is attributed to incorporating both autocorexperts' knowledge and spatial autocorrelation in the model construction. This paper is differentiated from previous studies in the sense that it applied the bayesian kriging technique to estimate price for commercial lots and explicitly combined experts' knowledge with data. It is expected that the result of this paper would provide a useful guide for the circumstances under which property price has to be estimated reliably based on sparse transaction data.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.1
• /
• pp.27-40
• /
• 2013

This study is about development of technology system for an advanced deep-sea unmanned underwater robot of S. Korea analysed by the application of scenario planning. It was developed a 6000m class next-generation deep-sea unmanned underwater vehicle(or robot, UUV) system, soonly ROV 'Hemire' and Depressor 'Henuvy' in 2006 at S. Korea and motion control, adaptive control algolithm, a work-space manipulator control algolithm, especially the underwater inertial-acoustic navigation system robust to initial errors and sensor failures. But there are remained matters on position tracking of the USBL, inertial-acoustic navigation system, attitude sensor, designed sonar sensors. So this study suggest the new idea for settle the matters and then this idea help the development of the underwater inertial-acoustic navigation system robust to initial errors and sensor failures, such as acoustic signal drop-out, by modifying the error covariance of the failed sonar signal when drop-out occurs. As a result, the future policy for deep-sea unmanned underwater robot of S. Korea is to further spur the development of new technology and more improvement of the technology level for deep-sea unmanned underwater robot system with indicator and imaginary wall as external device.

• Environmental Analysis Health and Toxicology
• /
• v.29
• /
• pp.12.1-12.8
• /
• 2014

Objectives Cohort studies of associations between air pollution and health have used exposure prediction approaches to estimate individual-level concentrations. A common prediction method used in Korean cohort studies is ordinary kriging. In this study, performance of ordinary kriging models for long-term particulate matter less than or equal to $10{\mu}m$ in diameter ($PM_{10}$) concentrations in seven major Korean cities was investigated with a focus on spatial prediction ability. Methods We obtained hourly $PM_{10}$ data for 2010 at 226 urban-ambient monitoring sites in South Korea and computed annual average $PM_{10}$ concentrations at each site. Given the annual averages, we developed ordinary kriging prediction models for each of the seven major cities and for the entire country by using an exponential covariance reference model and a maximum likelihood estimation method. For model evaluation, cross-validation was performed and mean square error and R-squared ($R^2$) statistics were computed. Results Mean annual average $PM_{10}$ concentrations in the seven major cities ranged between 45.5 and $66.0{\mu}g/m^3$ (standard deviation=2.40 and $9.51{\mu}g/m^3$, respectively). Cross-validated $R^2$ values in Seoul and Busan were 0.31 and 0.23, respectively, whereas the other five cities had $R^2$ values of zero. The national model produced a higher cross-validated $R^2$ (0.36) than those for the city-specific models. Conclusions In general, the ordinary kriging models performed poorly for the seven major cities and the entire country of South Korea, but the model performance was better in the national model. To improve model performance, future studies should examine different prediction approaches that incorporate $PM_{10}$ source characteristics.

• Journal of Korean Society of Forest Science
• /
• v.78 no.3
• /
• pp.323-330
• /
• 1989

Pinass densiflora natural forests of four local forms in Korea were studies to investigate effective biomass estimation method. Dimension analysis was used and three allometric regression models, such as logWt=A+BlogD, logWt=$A+B1ogD^2H$ and 1ogWt=A+BlogD+ClogH were applied to estimate biomass, The most accurate estimation was made by the regression model of logWt=A+BlogD+ClogH where Wt is dry weight, D is diameter at breast height, and H is tree height. However, dry weights of cones and dead branches were remotely related to tree size factor, such as D and H. In the interest of practical use. generalized allometric regressions for all samples trees of four stands were computed and analysis of covariance was used to compare the allometric regressions among the four stands. Based on the test criteria applied in this study, significant differences were found in terms of error variance and regression intercept, not in terms of regression slope. These trends suggest a generalized biomass regression is not valid for accurate estimation over a range of four local form stands.

• Animal Bioscience
• /
• v.35 no.10
• /
• pp.1499-1511
• /
• 2022

Objective: This study was aimed to estimate the genetic parameters, including genetic and phenotypic correlations, of milk yield, lactation curve traits and milk composition of Thai dairy cattle from three government research farms. Methods: The data of 25,789 test-day milk yield and milk composition records of 1,468 cattle from lactation 1 to 3 of Holstein Friesian (HF) and crossbred HF dairy cattle calved between 1990 and 2015 from three government research farms in Thailand were analysed. 305-day milk yield was estimated by the Wood model and a test interval method. The Wood model was used for estimating cumulative 305-day milk yield, peak milk yield, days to peak milk yield and persistency. Genetic parameters were estimated using linear mixed models with herd, breed group, year and season of calving as fixed effects, and animals linked to a pedigree as random effects, together with a residual error. Univariate models were used to estimate variance components, heritability, estimated breeding values (EBVs) and repeatability of each trait, while pairwise bivariate models were used to estimate covariance components and correlations between traits in the same lactation and in the same trait across lactations. Results: The heritability of 305-day milk yield, peak milk yield and protein percentage have moderate to high estimates ranging from 0.19 to 0.45 while days to peak milk yield, persistency and fat percentage have low heritability ranging from 0.08 to 0.14 in lactation 1 cows. Further, heritability of most traits considered was higher in lactation 1 compared with lactations 2 and 3. For cows in lactation 1, high genetic correlations were found between 305-day milk yield and peak milk yield (0.86±0.07) and days to peak milk yield and persistency (0.99±0.02) while estimates of genetic correlations between the remaining traits were imprecise due to the high standard errors. The genetic correlations within the traits across lactation were high. There was no consistent trend of EBVs for most traits in the first lactation over the study period. Conclusion: Both the Wood model and test interval method can be used for milk yield estimates in these herds. However, the Wood model has advantages over the test interval method as it can be fitted using fewer test-day records and the estimated model parameters can be used to derive estimates of other lactation curve parameters. Milk yield, peak milk yield and protein percentage can be improved by a selection and mating program while days to peak milk yield, persistency and fat percentage can be improved by including into a selection index.

• Journal of Radiation Industry
• /
• v.17 no.1
• /
• pp.19-32
• /
• 2023

The amount of radioactive waste is expected to dramatically increase with decommissioning of nuclear power plants such as Kori-1, the first nuclear power plant in South Korea. Accurate nuclide analysis is necessary to manage the radioactive wastes safely, but research on verification of radionuclide analysis has yet to be well established. This study aimed to develop the technology that can verify the results of radionuclide analysis based on artificial intelligence. In this study, we propose an anomaly detection algorithm for inspecting the analysis error of radionuclide. We used the data from 'Updated Scaling Factors in Low-Level Radwaste' (NP-5077) published by EPRI (Electric Power Research Institute), and resampling was performed using SMOTE (Synthetic Minority Oversampling Technique) algorithm to augment data. 149,676 augmented data with SMOTE algorithm was used to train the artificial neural networks (classification and anomaly detection networks). 324 NP-5077 report data verified the performance of networks. The anomaly detection algorithm of radionuclide analysis was divided into two modules that detect a case where radioactive waste was incorrectly classified or discriminate an abnormal data such as loss of data or incorrectly written data. The classification network was constructed using the fully connected layer, and the anomaly detection network was composed of the encoder and decoder. The latter was operated by loading the latent vector from the end layer of the classification network. This study conducted exploratory data analysis (i.e., statistics, histogram, correlation, covariance, PCA, k-mean clustering, DBSCAN). As a result of analyzing the data, it is complicated to distinguish the type of radioactive waste because data distribution overlapped each other. In spite of these complexities, our algorithm based on deep learning can distinguish abnormal data from normal data. Radionuclide analysis was verified using our anomaly detection algorithm, and meaningful results were obtained.

• Journal of Animal Science and Technology
• /
• v.51 no.5
• /
• pp.345-352
• /
• 2009

This study utilized 332,625 records of linear type scores consisting for 15 primary traits, 22,175 final score and 84,612 pedigree information of 22,175 Holstein cows from 1993 to 2007 in Korea to estimate genetic parameters for 16 type traits. Genetic and error (co)variances between two traits selected from 16 traits were estimated using bi-trait pairwise analyses with DFREML package. The estimated heritabilities for stature (ST), strength (STR), body depth (BD), dairy form (DF), rump angle (RA), thurl width (TW), rear legs side view (RLSV), foot angle (FA), fore udder attachment (FUA), rear udder height (RUH), rear udder width (RUW), udder cleft (UC), udder depth (UD), front teat placement (FTP), front teat length (FTL) and final score (FS) were 0.31, 0.21, 0.25, 0.10, 0.29, 0.19, 0.09, 0.06, 0.12, 0.13, 0.12, 0.08, 0.26, 0.20, 0.28 and 0.15, respectively. ST had the highest positive genetic correlation with BD (0.90), while RLSV had the highest negative genetic correlation with FA (-0.56). RA had negative genetic correlation with most udder traits (-0.17~-0.02). Especially, RUW had the higher positive genetic correlation with STR (0.60), BD (0.62), and TW (0.49), however, UD had the higher negative genetic correlation with STR (-0.40) and BD (-0.40). FTL had negative genetic correlation with FUA, RUH, RUW, UC and UD. FS had positive genetic correlation with UC, UD and FTP (0.12, 0.18 and 0.20). However, additional research is needed on the use of these parameters in the genetic evaluation because estimated genetic and error variance-covariance matrices were not positive definite.