• Korean Journal of Weed Science
• /
• v.30 no.1
• /
• pp.50-57
• /
• 2010

This experiment was conducted to investigate seedling emergence of Humulus japonicus in Seoul and Suwon and to predict its seedling emergence in various locations in Korea. Non-linear regression analysis to fit Gompertz model to accumulated seedling emergence of H. japonicus was performed to describe its seedling emergence. Parameter estimates from the non-linear regression and estimated effective accumulated temperature required for its seedling emergence were further used to predict seedling emergence of H. japonicus in 8 major geographical regions of Korea. Seedling emergence of H. japonicus versus effective accumulated temperature was well described by Gompertz model in both Seoul and Suwon. Effective accumulated temperatures required for the first seedling emergence and 50% of the maximum seedling emergence were estimated to be $56.7^{\circ}C$ and $76.3^{\circ}C$, respectively. Therefore, the dates for the first seedling emergence of H. japonicus were predicted to be 21 and 30 March in Daegu and Daejeon, while 4 and 6 April in Suwon and Chuncheon, respectively. The dates reaching 50% of maximum seedling emergence were also predicted to be 25 March in Daegu, while 3, 8 and 9 April in Daejeon, Suwon and Chuncheon, respectively.

• The KIPS Transactions:PartB
• /
• v.9B no.2
• /
• pp.139-146
• /
• 2002

The robot soccer simulation game is a dynamic multi-agent environment. In this paper we suggest a new reinforcement learning approach to each agent's dynamic positioning in such dynamic environment. Reinforcement learning is the machine learning in which an agent learns from indirect, delayed reward an optimal policy to choose sequences of actions that produce the greatest cumulative reward. Therefore the reinforcement learning is different from supervised learning in the sense that there is no presentation of input-output pairs as training examples. Furthermore, model-free reinforcement learning algorithms like Q-learning do not require defining or learning any models of the surrounding environment. Nevertheless these algorithms can learn the optimal policy if the agent can visit every state-action pair infinitely. However, the biggest problem of monolithic reinforcement learning is that its straightforward applications do not successfully scale up to more complex environments due to the intractable large space of states. In order to address this problem, we suggest Adaptive Mediation-based Modular Q-Learning (AMMQL) as an improvement of the existing Modular Q-Learning (MQL). While simple modular Q-learning combines the results from each learning module in a fixed way, AMMQL combines them in a more flexible way by assigning different weight to each module according to its contribution to rewards. Therefore in addition to resolving the problem of large state space effectively, AMMQL can show higher adaptability to environmental changes than pure MQL. In this paper we use the AMMQL algorithn as a learning method for dynamic positioning of the robot soccer agent, and implement a robot soccer agent system called Cogitoniks.

• KIPS Transactions on Software and Data Engineering
• /
• v.1 no.3
• /
• pp.177-186
• /
• 2012

In this paper, we propose a real-time approach for detecting and tracking a hand region by analyzing depth images. We build a hand model in advance. The model has the shape information of a hand. The detecting process extracts out moving areas in an image, which are possibly caused by moving a hand in front of a camera. The moving areas can be identified by analyzing accumulated difference images and applying the region growing technique. The extracted moving areas are compared against a hand model to get justified as a hand region. The tracking process keeps the track of center points of hand regions of successive frames. For this purpose, it involves three steps. The first step is to determine a seed point that is the closest point to the center point of a previous frame. The second step is to perform region growing to form a candidate region of a hand. The third step is to determine the center point of a hand to be tracked. This point is searched by the mean-shift algorithm within a confined area whose size varies adaptively according to the depth information. To verify the effectiveness of our approach, we have evaluated the performance of our approach while changing the shape and position of a hand as well as the velocity of hand movement.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.6 no.1
• /
• pp.85-97
• /
• 2003

This study was aimed at suggest a proper logging method of mountain forest using geographic information system(GIS) based on digital terrain model(DTM) in the National Forest at Mt. Kumsan in Namhae-gun, Gyungsangnam-do, which has about 2,948 ha in area. The areal percentage of 201 to 250m in the elevation was about 15.5%, elevation of 251 to 300m was 14.5%, and 78.75% for higher than 400m. The accumulated areal percentage of below 30% in the gradient was 17.2%, and 81.0% for steeper than 60%. The area for tractor skidding was 17.2%(511.7ha), the area for tractor attached winch skidding was 63.8%(1,896.3ha) and 18.4%(545.5ha) for cable yarding. It is important to choose the proper logging machines for timber harvesting. In general, the selection of logging operation system was affected several major environmental factors like as terrain conditions(slope gradient, slope length) and stand factors. The rate of middle slope gradients in terrain conditions showed higher than that of steep slope gradients in this area. Therefore, it considered that the logging operation system in this area could apply to tractor+winch operating machine according to terrain conditions.

• Korean journal of applied entomology
• /
• v.51 no.2
• /
• pp.91-97
• /
• 2012

The dispersion indices, spatial pattern and sampling plan for pink citrus rust mite (PCRM), Aculops pelekassi, monitoring was investigated. Dispersion indices of PCRM indicated the aggregated spatial pattern. Taylor's power law provided better description of variance-mean relationship than Iwao's patchiness regression. Fixed-precision levels (D) of a sequential sampling plan were developed using by Taylor's power law parameters generated from PCRM on fruit sample (cumulated number of PCRM in $cm^2$ of fruit). Based on Kono-Sugino's empirical binomial the mean density per $cm^2$ could be estimated from fruit ratio with more than 12 rust mites per $cm^2$: $ln(m)=4.61+1.23ln[-ln(1-p_{12})]$. To determine the optimal tally threshold, the variance (var(lnm)) for mean (lnm) in Kono-Sugino equation was estimated. The lower and narrow ranged change of variance for esimated mean showed at a tally threshold of 12. To estimate PCRM mean density per $cm^2$ at fixed precision level 0.25, the required sample number was 13 trees, 5 fruits per tree and 2 points per fruit (total 130 samples).

• Tunnel and Underground Space
• /
• v.23 no.5
• /
• pp.442-453
• /
• 2013

A large-scale high-temperature thermal energy storage(TES) was numerically modeled and the heat loss through storage tank walls was analyzed using a commercial code, FLAC3D. The operations of rock cavern type and above-ground type thermal energy storages with identical operating condition were simulated for a period of five consecutive years, in which it was assumed that the dominant heat transfer mechanism would be conduction in massive rock for the former and convection in the atmosphere for the latter. The variation of storage temperature resulting from periodic charging and discharging of thermal energy was considered in each simulation, and the effect of insulation thickness on the characteristics of heat loss was also examined. A comparison of the simulation results of different storage models presented that the heat loss rate of above-ground type TES was maintained constant over the operation period, while that of rock cavern type TES decreased rapidly in the early operation stage and tended to converge towards a certain value. The decrease in heat loss rate of rock cavern type TES can be attributed to the reduction in heat flux through storage tank walls followed by increase in surrounding rock mass temperature. The amount of cumulative heat loss from rock cavern type TES over a period of five-year operation was 72.7% of that from above-ground type TES. The heat loss rate of rock cavern type obtained in long-period operation showed less sensitive variations to insulation thickness than that of above-ground type TES.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.12B
• /
• pp.1183-1191
• /
• 2010

In this paper, I propose an adaptive OFDM CP length algorithm for in PLC systems for smart grid. The proposed scheme calculates the channel delay information at the CP controller of the receiver by taking correlation between a received data frame and the following delayed one. The CP controller, immediately, feeds back the channel delay information to the transmitter. Then, the transmitter adapts CP length for next data frame. As an impulsive noise model, Middleton Class A interference model was employed. The performance is evaluated in terms of packet data rate, cumulative packet data rate, and bit error rate (BER). The simulation results showed data gain (which is the amount of the reduced bits) gets larger as the number of packets increase, but the amount of data gain reduced as the number of branches ($N_{br}$) increase. In respects of BER for the cases $N_{br}$ is 3, 4, and 5, performance of the adaptive CP length algorithm and the fixed CP scheme are similar. Therefore, it is confirmed the proposed scheme achieved data rate increment without BER performance reduction compared to the conventional fixed CP length scheme.

• Korean Journal of Weed Science
• /
• v.30 no.2
• /
• pp.84-93
• /
• 2010

Petri dish and pot experiments were conducted to investigate germination and seedling emergence of Ammannia coccinea as influenced by environmental factors. The best germination of A. coccinea was obtained at $35/30^{\circ}C$ of temperature and 0 bar of osmotic potential, while no germination at temperatures of ${\leq}$ $15^{\circ}C$ and ${\geq}$ $40^{\circ}C$, osmotic potentials of ${\leq}$ -2.0 bar, or dark condition. The best seedling emergence was observed at $35/30^{\circ}C$, at which the first emergence of A. coccinea was observed at 7 days after sowing (DAS) with its maximum emergence reached at 10 DAS. No seedling emergence was observed at $15/10^{\circ}C$ with significant reduction at $40/35^{\circ}C$. Seedling emergence decreased with increasing soil depth, resulting in no seedling emergence at ${\geq}$ 3 cm. The Gompertz model well described the cumulative germination and seedling emergence of A. coccinea with time. Germination influenced by osmotic potential and seedling emergence influenced by soil burial depth were well described by the logistic model. Overall results indicate that A. coccinea is photoblastic and requires temperatures greater than $15^{\circ}C$, osmotic potential greater than -2.0 bar, and soil burial depth shallower than 3 cm for its germination and seedling emergence, which were faster than M. vaginalis but slower than E. crus-galli.

• 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 Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.15 no.5
• /
• pp.405-412
• /
• 2022

This paper studies the method of estimating power loss and classifying the factors for improving the power generation efficiency through O&M. It is installed under various climatic conditions worldwide, operational and maintenance technologies suitable for the characteristics of the installation location are required. Existing studies related to solar power generation efficiency have been actively quantifying the impact on short-term losses by environmental factors such as high temperature, dust accumulation, precipitation, humidity, and wind speed, but analysis of the overall impact from a long-term operation perspective is limited. In this study, the potential for efficiency improvement was analyzed by re-establishing a loss classification system according to the power flow of solar power to derive a comprehensive efficiency model for long-term operation and estimating power loss through a case study for each region where climate conditions are classified. As a result of the analysis, the average annual potential for improving soiling loss was 26.9%, Death Valley 7.2%, and Seoul 3.8%. Aging losses was 6.6% in the 20th year as a cumulative. The average annual potential due to temperature loss was 2.9 % for Doha, 1.9% for Death Valley, and 0.2% for Seoul.