In-Bok Lee

Finding patterns with variable length gaps or don’t cares

In this paper we have presented new algorithms to handle the pattern matching problem where the pattern can contain variable length gaps. Given a pattern P with variable length gaps and a text T our algorithm works in O(n + m + α log(max

NUMERICAL SIMULATION OF VENTILATION EFFICIENCIES OF NATURALLY VENTILATED MULTI-SPAN GREENHOUSES IN KOREA

_{\rm 1<={\it i}<={\it l}}

PIV VERIFICATION OF GREENHOUSE VENTILATION AIR FLOWS TO EVALUATE CFD ACCURACY

(bi–ai))) time where n is the length of the text, m is the summation of the lengths of the component subpatterns, α is the total number of occurrences of the component subpatterns in the text and ai and bi are, respectively, the minimum and maximum number of don’t cares allowed between the ith and (i+1)st component of the pattern. We also present another algorithm which, given a suffix array of the text, can report whether P occurs in T in O(m + α loglogn) time. Both the algorithms record information to report all the occurrences of P in T. Furthermore, the techniques used in our algorithms are shown to be useful in many other contexts.

A Wind Tunnel Study of Natural Ventilation for Multi-Span Greenhouse Scale Models Using Two-Dimensional Particle Image Velocimetry (PIV)

In this study, the ventilation efficiencies of naturally ventilated multi-span greenhouses typically used in Korea were analyzed considering structural types and natural climate factors such as wind direction. It was critical to set up identical, stable climate conditions for fair analysis and comparison of ventilation efficiencies in regards to different configurations. However, these ideal conditions do not exist in reality. Therefore, computational fluid dynamics (CFD) simulation, which is able to visualize the airflow quantitatively and qualitatively, was used in this study to reduce the effect of variations between greenhouses and locations. The CFD model used in this study had an established ±6.7% error. In addition, the tracer gas decay (TGD) method was used to analyze ventilation efficiencies quantitatively to overcome the limitations of traditional ventilation analysis. The TGD results were then connected to the CFD model using user-defined function (UDF) tools. The ventilation rates computed by the TGD method showed a 16.7% average difference compared to rates determined using a conventional method. When the wind direction was perpendicular to the greenhouse side vent with 2 m s-1 wind speed at 10 m height, overall ventilation rates computed by TGD were 1.30, 1.23, 1.02, and 0.74 AC (air exchange) min-1 in summer conditions for widespan, 1-2W, widewidth, and venlo type greenhouses, respectively. The average ventilation rates with the wind parallel to the side vent were respectively 32%, 52%, 30%, and 75% lower than when perpendicular to the side vent, and respectively 50%, 89%, 47% and 94% lower than with the wind 45° to the side vent. The results indicate that 1-2W and venlo type greenhouses were less influenced by wind direction, likely due to the effect of roof openings.

The Evaluation of Dentinal Tubule Occlusion by Desensitizing Agents: A Real-time Measurement of Dentinal Fluid Flow Rate and Scanning Electron Microscopy

A computational fluid dynamics (CFD) model was developed to investigate the natural ventilation of greenhouses. Before investigating its accuracy, a wind tunnel and particle image velocimetry (PIV) test was initially conducted to find the best experimental conditions and improve the PIV accuracy. After many trials of the PIV tests, good results were obtained in the PIV accuracy test, and the errors were shown to be mostly in the .2.1% range, with a maximum of 5.4%. A single-span, wide-span greenhouse, one of the most popular types of multi-span greenhouses in Korea, was used to get qualitative and quantitative airflow distribution in the greenhouse using PIV and CFD. To improve the CFD accuracy, the PIV- and CFD-computed airflows in the greenhouse were compared, particularly on the distribution, local air velocity, and turbulent intensity in the greenhouse. The quality of the mesh density and the design of the boundary condition, especially that of the wind velocity and turbulence profiles, were found to be very important for getting accurate results. Assuming the PIV results were accurate, the most accurate CFD results were obtained when using renormalization-group (RNG) k-. and Reynolds stress turbulence numerical models. The errors of the CFD-computed air velocity when using the RNG k-. and Reynolds stress models were -7.9% and -7.6%, respectively. However, the average errors of turbulent intensity when the RNG k-. and Reynolds stress turbulence numerical models were used were -37.5% and -43.2%, respectively.

Numerical prediction of fugitive dust dispersion on reclaimed land in Korea.

A two–dimensional particle image velocimetry (PIV) system installed in a large–sized wind tunnel was used to examine the natural ventilation of fully open roof and Venlo–type multi–span greenhouses. Since the PIV system has rarely been used for large–scale, low–turbulence airflow conditions, the PIV system was investigated to improve its performance and accuracy. The PIV accuracy tests were conducted without any greenhouse models in the wind tunnel because it was very difficult to insert the anemometer sensors in the small–scale greenhouse models, and it was also assumed that the sensor itself could affect natural airflow patterns inside the models. Good results were obtained in the PIV accuracy test, and the errors were shown to be mostly in the .5% range. Additionally, the natural airflow in 2– to 10–span greenhouses was investigated using the PIV system. The PIV results indicated that the PIV technique was an effective method for evaluating the relative performance of alternative designs of greenhouse and other structures with turbulent interior air distributions.

A fast algorithm for the generalized k -keyword proximity problem given keyword offsets

The aims of this study were to examine changes in dentinal fluid flow (DFF) during the application of a desensitizing agent and to compare the permeability reduction levels among different types of desensitizing agents. A cervical cavity was prepared for the exposure of cervical dentin on an extracted human premolar connected to a subnanoliter fluid flow measuring device under 20 cm of water pressure. The cavity was acid-etched with 32% phosphoric acid to make dentin highly permeable. The different types of desensitizing agents that were applied on the cavity were Seal&Protect as the light-curing adhesive type, SuperSeal and BisBlock as oxalate types, Gluma Desensitizer as the protein-precipitation type, and Bi-Fluoride 12 as the fluoride type. DFF was measured from the time before the application of the desensitizing agent throughout the application procedure to five minutes after the application. The characteristics of dentinal tubule occlusion of each desensitizing agent were examined by scanning electron microscopy. The DFF rate after each desensitizing agent application was significantly reduced when compared to the initial DFF rate before application for all of the desensitizing agents (p<0.05). Seal&Protect showed a greater reduction in the DFF rate when compared to Gluma Desensitizer and Bi-Fluoride 12 (p<0.05). SuperSeal and BisBlock exhibited a greater reduction in DFF rate when compared to Bi-Fluoride 12 (p<0.05). The dentin hypersensitivity treatment effects of the employed desensitizing agents in this study were confirmed through real-time measurements of DFF changes. The light-curing adhesive and oxalate types showed greater reduction in the DFF rate than did the protein-precipitation and fluoride types.

δ γ --- Parameterized Matching

The Saemangeum reclaimed land located on the western coast of Korea is one of the worlds largest reclamation projects, developing 40,100 ha of land area. After construction of sea dikes in 2006, the exposed land area has been increasing in relation to the water level height. Dust from the exposed land, containing high amounts of salt, disperses to nearby areas. The dust is harmful to human and animal health as well as to plant growth. Therefore, an estimation of fugitive dust dispersion is necessary for the development of plans to manage the problem. Field experiments are ideal for understanding aerodynamic phenomena. However, finding a correlation between weather conditions and dust dispersion is a difficult task due to limited measuring points. Moreover, weather conditions are unstable, unpredictable, and cannot be artificially controlled. Field experiments also involve high labor and time expenses. To overcome these limitations, a computational fluid dynamics (CFD) model has been developed to analyze dust dispersion phenomena, both quantitatively and qualitatively, according to the topography of the area, under various weather conditions. Effort was devoted to improving the accuracy of the CFD model by taking into account topographical design, mesh structure, turbulence models, particle generation, and other factors influencing the final solution. Computed results of the 3-D developed CFD model were compared against experimental data. Results showed an average error of -6.8%, which is within the acceptable range. CFD-computed vertical log-profiles of dust dispersion were similar to the vertical profiles presented by an earlier study. CFD results showed that dispersion of fugitive dust was mainly affected by particle size, wind speed, wind direction, and topography of the area. The estimated dispersion distance, measured at a height of 3 m for the 10 µm particles and a wind speed of 1.7 m s-1, was 3100 m. A dispersion distance of 6300 m was obtained when wind speed was 3.9 m s-1 for the 10 µm particles. This study showed that a CFD model can be effectively used to supplement field experiments when analyzing dispersion of fugitive dust.

DEVELOPMENT OF VERTICAL WIND AND TURBULENCE PROFILES OF WIND TUNNEL BOUNDARY LAYERS

When searching for information on the Web, it is often necessary to use one of the available search engines. Because the number of results are quite large for most queries, we need some measure of relevance with respect to the query. One of the most important relevance factors is the proximity score, i.e., how close the keywords appear together in a given document. A basic proximity score is given by the size of the smallest range containing all the keywords in the query. We generalize the proximity score to include many practically important cases and present an O(n log k)-time algorithm for the generalized problem, where k is the number of keywords and n is the number of occurrences of the keywords in a document.

STUDY OF INTERNAL CLIMATE OF NATURALLY AND MECHANICALLY VENTILATED BROILER HOUSES

This paper defines a new pattern matching problem by combiningtwo paradigms: Δ γ ---matching andparameterized matching. The solution is essentially obtained by acombination of bitparallel techniques and a reduction to a graphmatching problem. The time complexity of the algorithm isO (nm ), assuming text size n , patternsize m and a constant size alphabet.