论文部分内容阅读
对一栋多功能低层建筑平屋顶上的太阳能光伏板进行风压实测,根据风压时程特性选择适当的概率密度函数对各测点的风压系数进行非线性回归分析.运用6种概率密度函数对各测点的均值和极值(正极大值和负极小值)风压系数时程以及面积平均后的极值风压系数时程进行数值拟合,比较不同概率分布的拟合效果.回归分析结果表明:均值风压系数的概率分布近似无偏分布,t分布的拟合效果最好,其次是Logistic分布、正态分布、极值I型分布;负压极小值和正压极大值风压系数的概率分布分别为左偏分布和右偏分布,极值I型分布的拟合效果最好,其次是Lognormal分布和Gamma分布,而正态分布的拟合效果最差;最不利风压系数由极值负风压控制,即光伏板的风力由吸力主导;经面积平均后的最不利风压系数建议取值-2.3.
The wind pressure of a multi-functional low-rise building flat roof was measured by wind pressure, and the appropriate probability density function was chosen according to the time-history characteristics of wind pressure to conduct non-linear regression analysis on wind pressure coefficient of each measuring point. Six kinds of probability density Function is used to numerically fit the time series of the wind pressure coefficient and the time series of the extreme value (positive value and negative value) of each measurement point and the time series of the maximum value of the wind pressure coefficient after the area is averaged to compare the fitting results of different probability distributions. The results of regression analysis showed that the probability distribution of mean wind pressure coefficient is almost unbiased distribution, the fitting effect of t distribution is the best, followed by Logistic distribution, normal distribution and extreme I distribution. The negative pressure minimum and positive pressure The probability distributions of wind pressure coefficients of large values are left-deviation distribution and right-deviation distribution respectively. The fitting results of extreme I-distribution are the best, followed by Lognormal distribution and Gamma distribution, while the normal distribution has the worst fitting effect. The adverse wind pressure coefficient is controlled by the extreme negative wind pressure, that is, the wind power of the photovoltaic panel is dominated by suction. The value of the most unfavorable wind pressure coefficient after the average of the area is suggested to be -2.3.