• Journal of Industrial Technology
• /
• v.25 no.B
• /
• pp.157-164
• /
• 2005

Two methods, the bin averaging method and least square method, are often used in calibrating wind turbine test sites. The objective of this work was to determine a better method to predict the wind speed at wind turbine installing point. The calibration was done at the test site on a complex terrain located in Daegwallyeong, Korea. It was performed for two different cases based on the IEC 61400-12 power performance measurement standard. The wind speeds averaged for 10 minutes ranged between 4 m/s and 16 m/s. The wind-direction bins of each meteorological mast were 10 degrees apart, and only the bins having data measured for more than 24 hours were employed for the test site calibration. For both cases, the two methods were found to yield almost same results which estimated real wind speed very closely.

• Wind and Structures
• /
• v.29 no.2
• /
• pp.87-98
• /
• 2019

As a regenerated turbulent wind field process, wind tunnel test has proven to be a promising approach for investigating the transmission tower-line system (TTLS) performance in view of experimental scaled models design, simulation techniques of wind field, and wind induced responses subjected to typhoon. However, the challenges still remain in using various wind tunnels to regenerate turbulent wind field with considerable progress having been made in recent years. This review paper provides an overview of the state-of-the-art of the wind tunnel based on active or passive controlled simulation techniques. Specific attention and critical assessment have been given to: (a) the design of experimental scaled models, (b) the simulation techniques of wind field, and (c) the responses of TTLS subjected to typhoon in wind tunnel. This review concludes with the research challenges and recommendations for future research direction.

• Wind and Structures
• /
• v.36 no.6
• /
• pp.355-366
• /
• 2023

Although machine learning (ML) techniques have been widely used in various fields of engineering practice, their applications in the field of wind engineering are still at the initial stage. In order to evaluate the feasibility of machine learning algorithms for prediction of wind loads on high-rise buildings, this study took the exposure category type, wind direction and the height of local wind force as the input features and adopted four different machine learning algorithms including k-nearest neighbor (KNN), support vector machine (SVM), gradient boosting regression tree (GBRT) and extreme gradient (XG) boosting to predict wind force coefficients of CAARC standard tall building model. All the hyper-parameters of four ML algorithms are optimized by tree-structured Parzen estimator (TPE). The result shows that mean drag force coefficients and RMS lift force coefficients can be well predicted by the GBRT algorithm model while the RMS drag force coefficients can be forecasted preferably by the XG boosting algorithm model. The proposed machine learning based algorithms for wind loads prediction can be an alternative of traditional wind tunnel tests and computational fluid dynamic simulations.

• Journal of Biosystems Engineering
• /
• v.31 no.4 s.117
• /
• pp.355-362
• /
• 2006

A one-dimensional heat conduction problem in cylindrical coordinate system was solved using Crank-Nicolson finite difference method to predicting the temperature distribution in rice storage bin with wall insulator. The model can simulate the grain temperatures in insulated round bins using the input data of initial grain temperature. ambient air temperature, wind velocity, solar radiation on a horizontal surface, and thermal properties of grain, bin wall, wall insulator, insulator cover, and air. Temperatures were collected at the bin center, 0.65m in radial direction from the center, and near the bin wall in 2.7m diameter bin filled with rough rice to depth of 3.0m were used to validate the simulation model. Grain temperatures predicted by the model were in very good agreement with the measured temperatures. The residual mean square error between measured and predicted grain temperatures at the bin center was $1.38^{\circ}C$.

• Journal of the korean society of oceanography
• /
• v.32 no.3
• /
• pp.93-102
• /
• 1997

Coastal circulations during the (surface condition of an) idealized cold-air outbreak are numerically investigated with two-dimensional, non-hydrostatic model in which a constant bottom-slope exists. The atmospheric forcing during a cold-air outbreak is incorporated as the surface cooling and the wind stress. When the offshore angle of the wind-stress vector, defined as the angle measured from the alongshore axis, is smaller than 45 degrees, a strong downwelling circulation develops near the coast. A sharp density front, which separates the vertically homogeneous region from the offshore stratified region, is formed near the coast and propagates offshore with time. Onshore side of the density front, small-scale circulation cells which are aligned in the direction perpendicular to the bottom begin to develop as the near-coast homogeneous region broadens. The surface cooling enhances greatly the development of the surface mixed layer by convective motions due to hydrostatic instability. The convective motions reach far below the hydrostatically unstable layer which is attached to the surface. The small-scale circulation cells are appreciably modified by the convetion cell and the density front develops far offshore compared to the case of no surface cooling. As to the effect of the bottom slope, the offshore distance of the density front increases (decreases) as the bottom slope decreases (increases), which results from the fact that the onshore volume-transport (Ekman transport) of the low-density upper seawater remains almost constant when the wind-stress is maintained constant. It is shown that the bottom slope is an essential factor for the formation of both the density front and the alongshore current when the surface cooling is the only forcing.

• Wind and Structures
• /
• v.29 no.6
• /
• pp.417-430
• /
• 2019

To study the wake influence of an upstream bridge on the wind-resistance performance of a downstream bridge, two adjacent long-span cable-stayed bridges are taken as examples. Based on wind tunnel tests, the static aerodynamic coefficients and the dynamic response of the downstream bridge are measured in the wake of the upstream one. Considering different horizontal and vertical distances, the flutter derivatives of the downstream bridge at different angles of attack are extracted by Computational Fluid Dynamics (CFD) simulations and discussed, and the change in critical flutter state is further studied. The results show that a train passing through the downstream bridge could significantly increase the lift coefficient of the bridge which has the same direction with the gravity of the train, leading to possible vertical deformation and vibration. In the wake of the upstream bridge, the change in lift coefficient of the downstream bridge is reduced, but the dynamic response seems to be strong. The effect of aerodynamic interference on flutter stability is related to the horizontal and vertical distances between the two adjacent bridges as well as the attack angle of incoming flow. At large angles of attack, the aerodynamic condition around the downstream girder which may drive the bridge to torsional flutter instability is weakened by the wake of the upstream bridge, and the critical flutter wind speed increases at this situation.

• Proceedings of the KSR Conference
• /
• 2009.05a
• /
• pp.152-161
• /
• 2009

This thesis aims to report the result of an evaluation of the aerodynamic characteristic on the purchase conditional pantograph model that has developed for applying to the KTX-I. Development pantograph was derived the result of aerodynamic characteristic through a wind tunnel test. And design modification was proceeded to solve the difference of aerodynamic characteristic by pantograph's operating direction. The verification test and adjustment test about the modified pantograph's aerodynamic characteristic were progressed on the KTX-I. To be corresponded with requirements, the airfoil and spoiler were used.

• Wind and Structures
• /
• v.23 no.4
• /
• pp.301-311
• /
• 2016

The numerous benefits of Savonius turbine such as simple in structure, has appropriate self-start ability, relatively low operating velocity, water acceptance from any direction and low environmental impact have generated interests among researchers. However, it suffers from a lower efficiency compared to other types of water turbine. To improve its performance, parameters such flow pattern, pressure and velocity in different conditions must be analyzed. For this purpose, a detailed description on the flow field of various types of Savonius rotors is required. This article presents a numerical study on a nonlinear two-dimensional flow over a classic Savonius type rotor and a Benesh type rotor. In this experiment, sliding mesh was used for solving the motion of the bucket. The unsteady Reynolds averaged Navier-Stokes equations were solved for velocity and pressure coupling by using the SIMPLE (Semi-Implicit Method for Pressure linked Equations) algorithm. Other than that, the turbulence model using $k-{\varepsilon}$ standard obtained good results. This simulation demonstrated the method of the flow field characteristics, the behavior of velocity vectors and pressure distribution contours in and around the areas of the bucket.

• Journal of the Korean Society for Railway
• /
• v.19 no.1
• /
• pp.11-19
• /
• 2016

The flows around a high speed train are very important because they could affect the aerodynamic characteristics such as drag and acoustic noise. Especially the boundary layer of flows could represent the characteristic of flows around the high speed train. Most previous studies have focused on the boundary layer region along the train length direction for the side of the train and underbody. The measurement and analysis of the boundary layer for the roof side is also very important because it could determine the flow inlet condition for the pantograph. In this study, the roof boundary layer was measured with a 1/20 scaled model of the next generation high speed train, and the results were compared with full-scaled computational fluid dynamics results to confirm their validity. As a result, it was confirmed that the flow inlet condition for the pantograph is about 85% of the train speed. Additionally, the characteristics of the boundary layer, which increases along the train direction, was also analyzed.