• 대한기계학회논문집B
• /
• 제24권12호
• /
• pp.1615-1624
• /
• 2000

Algebraic Reynolds Stree (ARS) model is applied in order to analyze the turbulent flow of wall-attaching offset jet and to evaluate the predictability of model. The applied numerical schemes are the upwind scheme and the skew-upwind scheme. The numerical results show a good prediction in the first order calculations(i.e., reattachment length, mean velocity, pressure), however, slight deviations in the second order(i.e., kinetic energy and turbulence intensity). Comparing with the previous results using the k-$\varepsilon$ model, the ARS model predicts better than the standard k-$\varepsilon$ model, however, slightly worse than the k-$\varepsilon$ model including the streamline curvature modification. Additionallay this study can reconfirm that the skew-upwind scheme has approximately 25% improved predictability than the upwind scheme.

• 한국막학회:학술대회논문집
• /
• 한국막학회 1998년도 추계 총회 및 학술발표회
• /
• pp.15-18
• /
• 1998

Membrane reactors are systems which combine a chemical reactor with a membrane separation process allowing to carry out simultaneously conversion and product separation. The catalyst can be immobilized on the membrane or simply compartmentalized in a reaction space by the membrane. Membrane reactors are today investigated to produce optically pure isomers and/or resolve racemic mixture of enantiomers. The interest towards these systems is due to the increasing demand of enantiomerically pure compounds to be used in the pharmaceutical, food, and agrochemical industries. In fact, enantiomers can have different biological activities, which often influence the efficacy or toxicity of the compound. On the basis of current literature there are basically two schemes on the use of membrane technology to produce enantiomers. In one case, the membrane itseft is intrinsically enantioselective: the membrane is the chiral system which selectively separates the wanted isomer on the basis of its conformation. In the other, a kinetic resolution using an enantiospecific biocatalyst is combined with a membrane separation process; the membrane separates the product from the substrate on the basis of their relative chemical properties (i.e. solubility). This kind of configuration is widely used to carry out kinetic resolutions of low water soluble substrams in biphasic membrane reactors [Giomo, 1995, 1997; Lopez, 1997]. These are systems where enzyme-loaded membranes promote reactions between two separate phases thanks to the properties of enzymes, such as lipases, to catalyse reactions at the org ic/aqueous interface; the two phases are maintained in contact and separated at the membrane level by operating at appropriate transmembrane pressure. A schematic representation of biphasic membrane reactor is shown in figure 1, while an example of enantiospecific reaction and product separation carried out with these systems is reported in figure 2.

• 설비공학논문집
• /
• 제7권4호
• /
• pp.622-632
• /
• 1995

The present numerical study is aimed to investigate time-dependent characteristics of a two-dimensional lid-driven square cavity flow of three high Reynolds numbers, $7.5{\times}10^3$, $10^4$ and $3{\times}10^4$. A conservative convection term on irregular grids was adopted by renewing the MAC type difference schemes on regular grids. Relaxation of velocity and pressure is implemented by SOLA algorithm. In case of $Re=7.5{\times}10^3$, flow behavior converges to steady state after a transient period. But for $Re=10^4$, periodic unsteady sinusoidal fluctuation of local velocity and kinetic energy is found and continuous movements of small eddies in the secondary flow regions are also discovered. Random generation of eddies and their active migrating behavior are detected for $Re=3{\times}10^4$, resulting in complete unsteady and non-linear flow characteristics. And, an organized structure similar to a Moffat vortex is also observed from the time-mean flow patterns. Furthermore, a typoon-like vortex(TLV) appears intemittently and rotates along the separation regions and boundary layers.

• Journal of Electrical Engineering and Technology
• /
• 제13권1호
• /
• pp.60-67
• /
• 2018

The presence of induction motor loads in a power system may cause the phenomenon of delayed voltage recovery after the occurrence of a severe fault. A high proportion of induction motor loads in the power system can be a significant influence on the voltage stability of the system. This problem referred to as FIDVR(Fault Induced Delayed Voltage Recovery) is commonly caused by stall of small HVAC unit(Heating, Ventilation, and Air Conditioner) after transmission or distribution system failure. This delayed voltage recovery arises from the dynamic characteristics associated with the kinetic energy of the induction motor load. This paper proposes the UVLS (Under Voltage Load Shedding) control strategy for dealing with FIDVR. UVLS based schemes prevent voltage instability by shedding the load and can help avoid major economic losses due to wide-ranging cascading outages. This paper review recent topic about under voltage load shedding and compare decentralized load shedding scheme with conventional load shedding scheme. The load shedding strategy is applied to an actual system in order to verify the proposed FIDVR mitigation solution. Simulations demonstrate the effectiveness of the proposed method in resolving the problem of delayed voltage recovery in the Korean Power System.

• 한국안전학회지
• /
• 제8권4호
• /
• pp.175-182
• /
• 1993

Volatilization of toxic heavy metals, especially, metal chlorides at elevated temperatures in oxidation conditions was observed using a thermogravimetric furnace since such metal chlorides used to be a cause for the disease of industrial workers by their toxicity and high volatile extent. Most of tested metal chloride compounds were evaporated or decomposed into gas phase at elevated temperatures ranged from 200~90$0^{\circ}C$, while CrCl$_3$ and NiC1$_2$became stable with converting into oxide forms. A kinetic model for evaporation/condensation could predict maximum evaporation flux and the calculated values were compared with real evaporation flux. The ratio of two fluxes could be explained as the fraction of impinging gas molecules to the condensing surface( $\alpha$ ) and obtained in the range of 10$^{-3}$ ~10$^{-9}$ for the experimented toxic heavy metal chlorides. This ratio might be used to define the volatile extent or toxicity of such toxic metal compounds. The schemes to avoid volatilization of toxic heavy metals Into the atmosphere were suggested as follows ; 1 ) controlling the compositions of metals and Chlorine produced substances( such as PVC ) in the treated materials using a reverse estimation from regulatory limit and characteristics of a processing facility, 2) Installation of wet type devices such as a scrubber for condensing the metal compounds.

• 한국농공학회논문집
• /
• 제58권6호
• /
• pp.55-69
• /
• 2016

RUSLE (Revised Universal Soil Loss Equation) is the empirical formular widely used to estimate rates of soil erosion caused by rainfall and associated overland flow. Among the factors considered in RUSLE, rainfall erosivity factor (R factor) is the major one derived by rainfall intensity and characteristics of rainfall event. There has been developed various methods to estimate R factor, such as energy based methods considering physical schemes of soil erosion and simple methods using the empirical relationship between soil erosion and annual total rainfall. This study is aimed to quantitatively evaluate the variation among the R factors estimated using different methods for South Korea. Station based observation (minutely rainfall data) were collected for 72 stations to investigate the characteristics of rainfall events over the country and similarity and differentness of R factors calculated by each method were compared in various ways. As results use of simple methods generally provided greater R factors comparing to those for energy based methods by 76 % on average and also overestimated the range of factors using different equations. The variation coefficient of annual R factors was calculated as 0.27 on average and the results significantly varied by the stations. Additionally the study demonstrated the rank of methods that would provide exclusive results comparing to others for each station. As it is difficult to find universal way to estimate R factors for specific regions, the efforts to validate and integrate various methods are required to improve the applicability and accuracy of soil erosion estimation.