The effect of two- and three-dimensional double steep hills on the mean velocity and turbulent characteristics of a boundary-layer were investigated in a wind tunnel. The mean flow and turbulence were measured using split-fiber probes designed for measuring flows with high turbulence and separation. The results were as follows: 1) Profiles of mean velocity and the standard deviation of the fluctuating velocity components above an upwind hill for continues two- and three-dimensional hills were similar to those of the the single hill. 2) The mean velocity data at a downwind hilltop for continues two- and three-dimensional hills had lower values compared to that of upwind hilltop data. However, the standard deviation of fluctuating velocity components on the downwind hilltop data was higher than the values of the upwind hilltop data. 3) The reattachment behind downwind hill for continues two- and three-dimensional hills was smaller than that of the single hill. 4) Velocity variance on downwind hilltop was greatly influenced by the distance between the double hill. The perturbation increased with the distance and the maximum perturbation in two-dimensional hills occurred beyond 12-15 hill heights, because of vortex shedding by upwind hill.
We have developed a magnetic suspension and balance system (MSBS) having a test section of 36cm×40cm and being controlled around 5 axes, which can be operated with low electric power. We have developed the control method to support the model in the center of the test section. In order to evaluate the characteristics of this MSBS, we performed the calibration tests of magnetic forces on five axes and interference between these axes. We measured drag coefficients of a sphere and support interference in order to confirm the performance of this MSBS.
Drag coefficients of a sphere were measured in support interference free by the NAL 60cm MSBS(Magnetic Suspension and Balance System) up to 3.5×105 in Reynolds number. The magnetically supported sphere moved a little around tunnel center during tests. The displacement parallel to the flow is about 1/4 times as small as perpendicular one. The displacement perpendicular to the flow was distributed uniformly in direction at speed of 20m/s.The obtained drag coefficients show almost the same values as the other source data sets using a MSBS. In unsteady aerodynamic force on the sphere, rms values of the side force were about 4 times as large as parallel one at speed of 35m/s. Two kinds of averaged side force were observed at Reynolds numbers not less than 2.5×105 while the same drag coefficient was measured. One increased rapidly with Reynolds number. The other keeps zero or very small magnitude as Reynolds number increases. Strouhal numbers were measured at three Reynolds numbers up to 2.8×105 and they show good agreement with ones in other sources.
Transmission towers are considered to be designed with enough wind-resistant performance, however, it is necessary to re-examine the current estimation methods from both points of design wind speed and wind load to rationalize the tower design and to improve the structural integrity. Wind resistant rationalization committee (FY1999-FY2001), organized by Central Research Institute of Electric Power Industry (CRIEPI), promoted the studies on directional basic wind speed and equivalent static wind load estimation following the previous committee on wind resistant design method for local winds effected by topography (FY1992-FY1998). The purpose of this paper is to describe the summary of the investigation results of the wind resistant rationalization committee and to show the outline of "Recommendations for wind loads on transmission towers - a draft (2002)".
Gust loading factor (GLF) method has been adopted to wind loads on buildings in most domestic and international codes or standards. The GLF consist of static, background and resonance factor. In the traditional GLF method, the background and resonance components have simply same vertical distribution of static one. But obviously, the each distribution of the components is different from the others. In this paper, two investigation related to wind load distribution have been carried out. Firstly, the equivalent static distribution of background component is proposed, which has agreement in shear force or over turning moment at any height of buildings. The distribution are derived by statistical approach, and verified by the wind tunnel test data. The other hand, distribution similar to static loads is inspected. Secondarily, the contributions of high natural frequency components are investigated by response analysis of typical structural model. As the result, the equivalent resonance load considered all modes at the building top amount 1.4 times as much as that only considered first mode, typically.
Drag coefficients of axial cylinders of 5 and 4.1 fineness ratios were measured in support interference free by the NAL 60cm MSBS(Magnetic Suspension and Balance System) at speeds from 10m/s to 35m/s. The drag coefficient of the cylinder of 5 fineness ratio is 0.891 at Reynolds number of 1.0×105 based on model diameter. It hardly varies with Reynolds number in range from 8×104 to 1.4×105. The drag coefficient of the cylinder of 4.1 fineness ratio is 0.867 at Reynolds number of 1.0×105. Contrary to the former case, it increases a little with Reynolds number in range from 5.0×104 to 1.1×105. Wake of the cylinder of 5 fineness ratio was measured with a pitot tube at 4.2 model diameters downstream at several flow speeds. The wake diameter (99% dynamic pressure ratio) is about 3 times as much as the model one in the Reynolds number range.