Research On The Effect Of Several Factors On The Wind Tunnel Test Results Of Bridge And Building Structure

Important civil engineering structures such as long-span bridges, high-rise building andlong-span space structures are located at the atmospheric boundary layer. Wind tunnel test isrecognizes that the most effective method to solve the wind engineering problem of importantengineering project at present. The study of wind effects of important engineering projectinduced by the atmospheric boundary layer still dependent on the wind tunnel test at presentand a long period in future. How to improve reliability and stability of the wind tunnel testresults is an important issue concerned by the researcher of wind engineering. Therefore, inorder to improve the research level of wind effects of structure and refine research theimportant engineering project, the basic issue of wind tunnel test techniques and the rules ofdifferent parameters influence on the wind tunnel test results and the error correction methodshould be studied systematically and deeply.This paper first research on the effect of several factors on the wind tunnel test results ofbridge and building structure. Among these, the following factors have been selected tospecially research。Sixteen different grids have been used to simulate the local turbulent flowfield, and summarize the rules of width and space of grids influence on the turbulenceintensity, turbulence integral scale and fluctuating wind power spectra at different wind-tunnelcross section. According to the results of wind field test, choosing proper wind characteristicparameters study the rules of turbulence intensity and turbulence integral scale influence onthe wind pressure distribution on different scale of bridge section and rectangular buildingstructure. Wind tunnel test is carried out to study the Jensen number effects on the rectangularbuilding structure, and summarize the rules of Jensen number effects influence on the meanwind load, fluctating wind load and its frequency characteristics of rectangular buildingstructure. Bridge sections with attachment structure (railing and track of inspection vehicle)have been made to study the attachment structure influence on the Reynolds number effects ofbridge section. The factors of two dimensional end plate and length-width ratio influence onthe wind tunnel test results of different kinds of bridge have been studied through wind tunneltest of section model. According to the above tests, simulation rules of the local turbulent flowfield and the rules of different parameters influence on the results of wind tunnel test of bridgeand building structure are summarized. The results show that:(1) When the size of the square hole in the center formed by grids fixed, turbulenceintensity increase with the increase of the grid’s width. Narrower the width of grids, lower theturbulence intensity, and better the data stability. The variation of the turbulence integral scale is not obvious compare with the turbulence intensity, but basically accord with the law ofincrease with the increase of the grid’s width. Larger turbulence integral scale would appearwhen the size of the square hole in the center formed by grids is30cm to40cm and the ratioof grid’s width to the size of square hole is0.5. The largest turbulence integral scale is close tothe size of square hole. An empirical formula is for evaluating the turbulence intensity andturbulence integral scale through the grid’s width and the size of the square hole in the centerformed by grids has been obtained.(2) The mean wind pressure on the negative pressure region of square building structureand bridge decrease with the increase of turbulence intensity, and the fluctuating windpressure increase with the increase of turbulence intensity. The mean wind pressure andfluctuating wind pressure of square building structure change with the turbulence intensityshow regular variation, and can be estimated by empirical formula. Turbulence intensityincrease10%, the mean wind pressure on windward side change10%—16%, broadsidechange16%—20%and leeward side change about20%. The mean wind pressure andfluctuating wind pressure of bridge and other complex shape structure change with theturbulence intensity can be influenced by many factors, such as wind directions, wind attackangles and signature turbulence, so it is difficult to estimate. The absolute value of mean windpressure increase with the increase of turbulence integral scale, but the amplitude changesrandomly. The fluctuating wind pressure has little influence by turbulence integral scale.(3) The impacts of Jensen number change on the fluctuating wind load of square buildingstructure can’t be ignored. Jensen number reduced from6000to1200, along fluctuating windload, across fluctuating wind load and torsional fluctuating wind load of square buildingstructure increase significantly, increase about2times to2.5times. There is little impact ofJensen number change on along mean wind load, across mean wind load and torsional meanwind load of square building structure, the max amplitude less than30%.(4) If the mean pressure of structure surface is positive, it is increase with the increase ofReynolds number. If the mean pressure is negative, it is decrease with the increase ofReynolds number. The fluctuating wind pressure will decrease with the increase of Reynoldsnumber. Larger size of turbulence integral scale leads to the Reynolds number effects of themean wind pressure and the fluctuating wind pressure on broadside and leeward side will bemore obvious. The influence of Reynolds number on the different position is different.Sensitivity of different degree appeared in the law of tri-component force varying withReynolds number to railing and track of inspection vehicle. Pressure measurement test in thewind tunnel should avoid the critical zone of Reynolds number as far as possible. The Reynolds number effects of mean and fluctuating wind pressure coefficient decreased in thesupercritical zone.(5) The appropriate length-width ratio of force measurement wind tunnel test of sectionmodel should be greater than2:1. The larger of the length-width ratio, the more stable theresults and the better reliability. But the spanwise correlation of different sections coulddecrease when the length-width ratio is greater than2:1. In order to ensure the better spanwisecorrelation, the best length-width ratio of section model should be greater than2:1and lessthan4:1. The three-component coefficients’sensitivity to length-width ratio has significantlygreater than the two dimensional end plate. The larger of the length-width ratio, the smallerimpaction of the two dimensional end plate.