广西大学学报（自然科学版）
廣西大學學報（自然科學版）
엄서대학학보（자연과학판）
JOURNAL OF GUANGXI UNIVERSITY （NATURAL SCIENCE EDITION）
2015年
4期
856-861
,共6页
Bootstrap方法%平均风剖面%现场实测%风荷载
Bootstrap方法%平均風剖麵%現場實測%風荷載
Bootstrap방법%평균풍부면%현장실측%풍하재
Bootstrap method%mean wind speed profile%field measurement%wind loads
通过选取近海岸100 m测风塔获取的10 m、50 m、65 m、80 m、100 m等5个高度台风和季风风场强风样本,研究强风条件下近地台风和季风风场的平均风剖面参数合理取值,应用Bootstrap统计分析方法,分析了强风样本容量较小条件下平均风剖面参数变异性及置信区间。分析结果表明：近地台风风场平均风速剖面符合指数律或对数律,台风风场风剖面指数均值为0.12,95%置信度条件下的置信区间为[0.117,0.125]；季风风场风剖面指数均值为0.07,95%置信度条件下的置信区间为[0.065,0.070]；台风和季风风场平均风剖面指数均值估计的概率分布呈高斯分布特征,均值的置信区间相对较窄、趋于稳定,分析结果具有较好的精度和可信度。
通過選取近海岸100 m測風塔穫取的10 m、50 m、65 m、80 m、100 m等5箇高度檯風和季風風場彊風樣本,研究彊風條件下近地檯風和季風風場的平均風剖麵參數閤理取值,應用Bootstrap統計分析方法,分析瞭彊風樣本容量較小條件下平均風剖麵參數變異性及置信區間。分析結果錶明：近地檯風風場平均風速剖麵符閤指數律或對數律,檯風風場風剖麵指數均值為0.12,95%置信度條件下的置信區間為[0.117,0.125]；季風風場風剖麵指數均值為0.07,95%置信度條件下的置信區間為[0.065,0.070]；檯風和季風風場平均風剖麵指數均值估計的概率分佈呈高斯分佈特徵,均值的置信區間相對較窄、趨于穩定,分析結果具有較好的精度和可信度。
통과선취근해안100 m측풍탑획취적10 m、50 m、65 m、80 m、100 m등5개고도태풍화계풍풍장강풍양본,연구강풍조건하근지태풍화계풍풍장적평균풍부면삼수합리취치,응용Bootstrap통계분석방법,분석료강풍양본용량교소조건하평균풍부면삼수변이성급치신구간。분석결과표명：근지태풍풍장평균풍속부면부합지수률혹대수률,태풍풍장풍부면지수균치위0.12,95%치신도조건하적치신구간위[0.117,0.125]；계풍풍장풍부면지수균치위0.07,95%치신도조건하적치신구간위[0.065,0.070]；태풍화계풍풍장평균풍부면지수균치고계적개솔분포정고사분포특정,균치적치신구간상대교착、추우은정,분석결과구유교호적정도화가신도。
Mean wind speed profile parameters of strong winds in surface boundary layer were stud-ied by the bootstrap method based on small wind data samples which were recorded from a 100 m meteorological tower at the site near seashore under strong winds like typhoons and monsoon. The variation of mean wind speed profile parameter and the relationship between mean wind speed profile parameter and sample size with the confidence interval were analyzed. The results reveal that, in the near-surface range (<100 m) , the vertical distribution of mean wind speed can be well described by a logarithmic law and a power law. The mean values of exponent of power-law profile were 0. 12 and 0. 07 in the typhoons and monsoon wind climates, respectively for the coastal terrain. Moreover, its confidence interval was [ 0 . 117 , 0 . 125 ] and [ 0 . 067 , 0 . 074 ] respectively at the confidence of 95%. It is found from the analysis the probability distribution of power-law index was found to follow the Gaussian distribution, and their confidence interval of the mean values of exponent were relative-ly narrow and tended to be stable. The results rationally demonstrated the precision and credibility of the proposed method for the mean wind speed profile parameter based on small samples of strong wind speed.
