(1 同济大学土木工程防灾国家重点实验室, 上海 200092; 2 广东省公路勘察规划设计院有限公司, 广州 510507)

［摘要］基于有限的风洞试验测试结果，提出了结合本征正交分解（POD）和薄板样条插值（TPS）预测未知点风压的方法，并将此方法应用于一座实际大跨度屋盖结构，在时域和频域上预测值与试验结果的对比证明了该方法的有效性和实用性。预测值与实测值的自功率谱在低频及高频部分都吻合得较好；相比自功率谱，相干函数预测的精度要低一些。同时指出POD\|TPS法得到的风压值与直接薄板样条插值结果几乎是完全相同的，其预测结果完全取决于插值方法；POD法的好处在于，它将对时程的插值转换为对少量本征模态的插值，大大降低了计算量。

［关键词］POD；大跨度屋盖；风压；预测

中图分类号：TU393.3， TU312.1文献标识码：A文章编号：1002-848X(2011)06-0098\|05

Application of POD combined with thin\|plate splines in research on wind pressure

Zhou Xuanyi¹, Li Gang²(1 State Key Laboratory of Disaster Reduction in Civil Engineering, Tongji University, Shanghai 200092, China; 2 Guangdong Highway Design Institute Co., Ltd., Guangzhou 510507, China)

Abstract:The proper orthogonal decomposition (POD) method combined with thin plate splines(TPS) interpolation technique was used to predict the wind pressures on a large-span roof based on the limited data from the wind tunnel test. The validity and practicability of POD\|TPS method was illustrated by comparing the predicted results with original data in the time domain and frequency domain. The predicted results of auto\|power spectrum agrees well with original ones both in low and high frequency range, whereas the prediction accuracy for cross\|power spectrum is not better than that for auto\|power spectrum. At the same time, it is pointed out that, the predicted wind pressures obtained through POD\|TPS method are almost the same as that from TPS method, which indicates that the predicted results only depend on the interpolation technique. When adopting POD method, the interpolation of time series is replaced by that of space POD mode, which reduces the calculation cost.

Keywords:POD; large-span roof; wind pressure; prediction

*科技部国家重点实验室基金资助（SLDRCE10\|B\|04）及国家自然科学基金创新研究群体科学基金（50621062）联合资助。

作者简介：周晅毅，博士，副研究员，Email:zhouxytj@tongji.edu.cn。

参考文献

［1］  KWATRA N, GODBOLE P N, KRISHNA P. Application of artificial neural network for determination of wind induced pressures on gable roof［J］. Wind & Structure, 2002, 5(1): 1\|14.

［2］  CHEN Y, KOPP G A, SURRY D. Interpolation of wind\|induced pressure time series with an artificial neural network［J］. Journal of Wind Engineering and Industrial Aerodynamics, 2002, 90(6): 589\|615.

［3］  顾明, 周晅毅. 神经网络方法在大跨度屋面风压研究中的应用［J］. 工程力学, 2003, 20(4): 99\|103.

［4］  CHEN Y,  KOPP G A,  SURRY D. Interpolation of pressure time series in an aerodynamic database for low buildings［J］. Journal of Wind Engineering and Industrial Aerodynamics, 2003, 91(6): 737\|765.          

［5］  FU J Y, LIANG S G, LI Q S. Prediction of wind\|induced pressures on a large gymnasium roof using artificial neural networks［J］. Computers and Structures, 2007, 85(3\|4): 179\|192.

［6］  KHO S, BAKER C, HOXEY R. POD/ARMA reconstruction of the surface pressure field around a low rise structure［J］. Journal of Wind Engineering and Industrial Aerodynamics, 2002, 90(12\|15): 1831\|1842.

［7］  RUAN D, HE H, CASTANON D A, et al. Normalized proper orthogonal decomposition (NPOD) for building pressure data compression［J］. Journal of Wind Engineering and Industrial Aerodynamics, 2006, 94(6): 447\|461.

［8］  李元齐, 沈祖炎. 本征正交分解法在曲面模型风场重构中的应用［J］. 同济大学学报：自然科学版, 2006, 34(1): 22\|26.

［9］  倪振华, 江棹荣, 谢壮宁. 本征正交分解技术及其在预测屋盖风压场中的应用［J］. 振动工程学报, 2007, 20(1): 1\|8.

［10］  李方慧, 倪振华, 沈世钊. POD方法在双坡屋盖风压场预测中的应用［J］. 工程力学, 2007, 24(2): 68\|73, 79.

［11］  江棹荣, 倪振华, 谢壮宁. 本征正交分解技术在屋盖风压场重建与预测中的应用［J］. 应用力学学报, 2007, 24(4): 592\|598.

［12］  江棹荣, 倪振华, 谢壮宁. POD在大跨屋盖风致响应计算中的应用［J］. 土木工程学报, 2007, 40(6): 1\|6.

［13］  GB50009—2001 建筑结构荷载规范［S］. 2006年版.北京：中国建筑工业出版社, 2006.

［14］  叶孟洋, 顾明, 黄鹏. 大跨度干煤棚结构表面体型系数试验研究［J］. 结构工程师, 2007, 23(1)：62-66.

［15］  SMITH M J, HODGES D H, CESNIK C E S. Evaluation of computational algorithms suitable for fluid\|structure interactions［J］. Journal of Aircraft, 2000, 37(2): 282\|294.