• Journal of Korean Society of Forest Science
• /
• v.111 no.4
• /
• pp.603-612
• /
• 2022

The study aims were to evaluate a machine-learning, algorithm-based, forest biomass-estimation model to estimate subnational forest biomass and to comparatively analyze REDD+ forest reference emission levels. Time-series Landsat satellite imagery and ESA Biomass Climate Change Initiative information were used to build a machine-learning-based biomass estimation model. The k-nearest neighbors algorithm (kNN), which is a non-parametric learning model, and the tree-based random forest (RF) model were applied to the machine-learning algorithm, and the estimated biomasses were compared with the forest reference emission levels (FREL) data, which was provided by the Paraguayan government. The root mean square error (RMSE), which was the optimum parameter of the kNN model, was 35.9, and the RMSE of the RF model was lower at 34.41, showing that the RF model was superior. As a result of separately using the FREL, kNN, and RF methods to set the reference emission levels, the gradient was set to approximately -33,000 tons, -253,000 tons, and -92,000 tons, respectively. These results showed that the machine learning-based estimation model was more suitable than the existing methods for setting reference emission levels.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2013.02a
• /
• pp.550-551
• /
• 2013

Large-area RF-driven ion source is being developed at Germany for the heating and current drive of ITER plasmas. Negative hydrogen (deuterium) ion sources are major components of neutral beam injection systems in future large-scale fusion experiments such as ITER and DEMO. RF ion sources for the production of positive hydrogen ions have been successfully developed at IPP (Max-Planck- Institute for Plasma Physics, Garching) for ASDEX-U and W7-AS neutral beam injection (NBI) systems. In recent, the first NBI system (NBI-1) has been developed successfully for the KSTAR. The first and second long-pulse ion sources (LPIS-1 and LPIS-2) of NBI-1 system consist of a magnetic bucket plasma generator with multi-pole cusp fields, filament heating structure, and a set of tetrode accelerators with circular apertures. There is a development plan of large-area RF ion source at KAERI to extract the positive ions, which can be used for the second NBI (NBI-2) system of KSTAR, and to extract the negative ions for future fusion devices such as ITER and K-DEMO. The large-area RF ion source consists of a driver region, including a helical antenna (6-turn copper tube with an outer diameter of 6 mm) and a discharge chamber (ceramic and/or quartz tubes with an inner diameter of 200 mm, a height of 150 mm, and a thickness of 8 mm), and an expansion region (magnetic bucket of prototype LPIS in the KAERI). RF power can be transferred up to 10 kW with a fixed frequency of 2 MHz through a matching circuit (auto- and manual-matching apparatus). Argon gas is commonly injected to the initial ignition of RF plasma discharge, and then hydrogen gas instead of argon gas is finally injected for the RF plasma sustainment. The uniformities of plasma density and electron temperature at the lowest area of expansion region (a distance of 300 mm from the driver region) are measured by using two electrostatic probes in the directions of short- and long-dimension of expansion region.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.13 no.4
• /
• pp.293-297
• /
• 2020

This paper tried to find the best conditions that could be applied to solar cells by deposition of CdS thin film on ITO glass using multiplex displacement sputter system. RF power was changed to 50W, 100W, and 150W and sputtering time was set to 10 minutes. As a result of the measurement of transmittance, the average transmittance in the area of 400 to 800 nm was measured from 60% to 80% and the best characteristic was measured at 150W at 84%. The band gap was also measured at 3.762eV at 50W, 4.037eV at 100W and 4.052eV at 150W. In XRD analysis, even as RF power was increased, it was observed as a structure called Wurtzite (hexagonal) of CdS. And as RF power increased, the particles were large and uniformly deposited, but at 100W the particles were densely composed and dense. And the thickness measurement showed that the RF power increased uniformly.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.2203-2205
• /
• 2003

KIRAMS-13, the first medical cyclotron developed by domestic technique, is used to produce radio-isotope such as $^{18}F$ whose life time is relatively short through test operation. For high-power operation of charged particle accelerators, the power coupler must withstand enormous stresses due to charging induced by high RF power passing through. High-power RF testing with peak power in excess of 30kW has been performed on prototype power coupler for KIRAMS-13 normal conducting cavities. CST MICROWAVE STUDIO(CST MWS) is used for fundamental RF Design, and power coupler is manufactured according to fundamental power coupler design requisite. The qualification of the couplers has occurred for the time being only in a limited set of conditions as the available RF system and control instrumentation are under improvement.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.19 no.7
• /
• pp.624-630
• /
• 2006

ZnO deposition parameters are not independent and have a nonlinear and complex property. To propose a method that could verify and predict the relations of process variables, neural network was used. At first, ZnO thin films were deposited by using RF magnetron sputtering process with various conditions. Si, GaAs, and Glass were used as substrates. The temperature, work pressure, and RF power of the substrate were $50\sim500^{\circ}C$, 15 mTorr, and $180\sim210W$, respectively : the purity of the target was ZnO 4 N. Structural properties of ZnO thin films were estimated by using XRD (0002) peak intensity. The structure of neural network was a form of 4-7-1 that have one hidden layer. In training a network, learning rate and momentum were selected as 0.2, 0.6 respectively. A backpropagation neural network were performed with XRD (0002) peak data. After training a network, the temperature of substrate was evaluated as the most important parameter by sensitivity analysis and response surface. As a result, neural network could capture nonlinear and complex relationships between process parameters and predict structural properties of ZnO thin films with a limited set of experiments.

• Journal of the Korean Institute of Telematics and Electronics D
• /
• v.36D no.5
• /
• pp.54-61
• /
• 1999

We suggested a macro medel for MOS transistors, which incorporates the distributed substrate resistance by using a method which utilizes external diodes on SPICE MOS model. By fitting the simulated s-parameters to the measures ones, we obtained a model set for the W=200TEX>$\mu\textrm{m}$ and L=0.8TEX>$\mu\textrm{m}$ NMOS transistor, and also analyzed the effects of distributed substrate resistance in the RF range. By comparing the physical parameters calculated from simulated s-parameters such as ac resistances and capacitances with the measured ones, we confirmed the validity of the simulation results. For the frequencies below 10GHz, it seems appropriated to use a simple macro model which utilizes the existing SPICE MOS model with junction diodes, after including one lumped resistor each for gate and substrate nodes.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.9
• /
• pp.21-27
• /
• 2015

An improved method based on direct extraction and simultaneous optimization is developed to determine model parameters of symmetric inductors fabricated by the high resistivity(HR) silicon-on-insulator(SOI) RF CMOS process. In order to improve modeling accuracy, several model parameters are directly extracted by Y and Z-parameter equations derived from two equivalent circuits of symmetric inductor and grounded center-tap one, and the number of unknown parameters is reduced using parallel resistance and total inductance equations. In order to improve optimization accuracy, two sets of measured S-parameters are simultaneously optimized while same model parameters in two equivalent circuits are set to common variables.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.7 no.4
• /
• pp.107-117
• /
• 2008

OBE(On Board Equipment) of IR(Infrared)/RF(Radio Frequence)-DSRC(Dedicated Short Range Communication) are set up 1.25 million cars because Korea Highway Cooperation supplied in HiPass(Domestic Electronic Toll Collection System). But this DSRC system has strength and weakness of service for BIS(Bus Information System) or UTIS(Urban Traffic Information System) because BIS and UTIS are required vary high communication's performance and management. In this paper, Efficiency of IR/RF-DSRC is analyzed performance of MAC(Medium Access Control) by computer simulation for services of BIS or UTIS, so Suggested to be commensurated with medium of there service. Specially, DSRC is needed high mobility and throughput in the vary short communication range, so Analysed the transmission rate of data in each communication medium over the specific character of the MAC frame.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.05a
• /
• pp.896-898
• /
• 2012

Recently, many countries are making efforts for the development of ocean resources because the necessity and importance of the ocean resources are increased. Underwater sensor networks have emerged as a very powerful technique for many applications, including monitoring, measurement, surveillance and control and envisioned to enable applications for oceanographic data collection, ocean sampling, environmental and pollution monitoring, offshore exploration, disaster prevention, tsunami and seaquake warning, assisted navigation, distributed tactical surveillance, and mine reconnaissance. The idea of applying sensor networks into underwater environments (i.e., forming underwater sensor networks) has received increasing interests in monitoring aquatic environments for scientific, environmental, commercial, safety, and military reasons. The data obtained by observing around the environment are wireless-transmitted by a radio set with various waves. According to the technical development of the medium set, some parameters restricted in observing the ocean have been gradually developed with the solution of power, distance, and corrosion and watertight by the seawater. The actual matters such as variety of required data, real-time observation, and data transmission, however, have not enough been improved just as various telecommunication systems on the land. In this paper, a wireless management system will be studied through a setup of wireless network available at fishery around the coast, real-time environmental observation with RF sensor, and data collection by a sensing device at the coastal areas.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.5
• /
• pp.153-160
• /
• 2020

Energy Consumption Predictions for Industries has a prominent role to play in the energy management and control system as dynamic and seasonal changes are occurring in energy demand and supply. This paper introduces and explores the steel industry's predictive models of energy consumption. The data used includes lagging and leading reactive power lagging and leading current variable, emission of carbon dioxide (tCO2) and load type. Four statistical models are trained and tested in the test set: (a) Linear Regression (LR), (b) Radial Kernel Support Vector Machine (SVM RBF), (c) Gradient Boosting Machine (GBM), and (d) Random Forest (RF). Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) are used for calculating regression model predictive performance. When using all the predictors, the best model RF can provide RMSE value 7.33 in the test set.