基于相空间重构的神经网络风暴潮增水预测方法-海洋预报

基于相空间重构的神经网络风暴潮增水预测方法

单位：国家海洋环境预报中心国家海洋局海洋灾害预报技术研究重点实验室, 北京 100081

分类号：P731.23

出版年·卷·期（页码）：2016·33·第一期（59-64）

摘要：

风暴潮增水的准确预测对于国民生产、防灾减灾有重大意义。本文提出一种基于相空间重构的神经网络风暴潮增水预测方法,即使用单站风暴潮增水数据重构出与之相关的相空间,然后使用BP神经网络模型拟合该相空间的空间结构。将该模型用于库克斯港风暴潮增水预测, 结果表明:该模型应用在风暴潮增水时间序列的预测中是合理、可行的, 并具有较高的精度。此外,使用db10小波函数对原始余水位数据进行降噪处理可以显著地提高模型的预测精度。

Accurate prediction of storm surge is significantly meaningful to the national production and disaster prevention and mitigation. The storm surge prediction model of BP neural network based on phase space was purposed in this paper, through combining reconstruction phase space with BP neural network. A phase space is reconstructed with the storm surge data and fitted with BP neural network model. The model is used to predict storm surge in Cuxhaven. The result of calculation shows the model is feasible, reasonable and highly precise. The prediction accuracy can been markedly improved through data de-noising.

参考文献：

[1] Packard N H, Crutchfield J P, Farmer J D, et al. Geometry from a time series[J]. Physical Review Letters, 1980, 45(9): 712-716.
[2] Takens F. Detecting strange attractors in turbulence[M]//Rand D, Young L S. Dynamical Systems and Turbulence, Warwick. Berlin Heidelberg: Springer, 1981: 366-381.
[3] Farmer J D, Sidorowich J J. Predicting chaotic time series[J]. Physical Review Letters, 1987, 59(8): 845-848.
[4] 吴耿锋, 周佩玲, 储阅春, 等. 基于相空间重构的预测方法及其在天气预报中的应用[J]. 自然杂志, 1999, 21(2): 107-110.
[5] McNames J, Suykens J A K, Vandewalle J. Winning entry of the K. U. Leuven time-series prediction competition[J]. International Journal of Bifurcation and Chaos, 1999, 9(8): 1485-1500.
[6] 向小东, 郭耀煌. 基于混沌吸引子的时间序列预测方法及其应用[J]. 西南交通大学学报, 2001, 36(5): 472-475.
[7] Yankov D, DeCoste D, Keogh E. Ensembles of nearest neighbor forecasts[M]//Fürnkranz J, Scheffer T, Spiliopoulou M. Machine Learning: ECML 2006. Berlin Heidelberg: Springer, 2006: 545- 556.
[8] 张颖超, 刘玉珠. 基于相空间重构的神经网络月降水量预测方法[J]. 计算机仿真, 2014, 31(1): 352-355.
[9] Solomatine D P, Rojas C J, Velickov S, et al. Chaos theory in predicting surge water levels in the North Sea[C]//Proceedings of the 4th International Conference on Hydroinformatics. Iowa, USA, 2000.
[10] Velickov S. Nonlinear dynamics and chaos with applications to hydrodynamics and hydrological modelling[M]. CRC Press, 2004.
[11] Siek M, Solomatine D, Velickov S. Multivariate chaotic models vs neural networks in predicting storm surge dynamics[C]//Proceedings of IEEE International Joint Conference on Neural Networks, 2008. IJCNN 2008. Hong Kong: IEEE, 2008: 2112- 2119.
[12] Siek M, Solomatine D P. Nonlinear chaotic model for predicting storm surges[J]. Nonlinear Processes in Geophysics, 2010, 17(5): 405-420.
[13] 王海燕, 卢山. 非线性时间序列分析及其应用[M]. 北京: 科学出版社, 2006.
[14] 史珍, 丁瑞强, 李建平. 随机误差对混沌系统可预报性的影响[J]. 大气科学, 2012, 36(3): 458-470.
[15] Fraser A M, Swinney H L. Independent coordinates for strange attractors from mutual information[J]. Physical Review A, 1986, 33(2): 1134-1140.
[16] 杨志安, 王光瑞, 陈式刚. 用等间距分格子法计算互信息函数确定延迟时间[J]. 计算物理, 1995, 12(4): 442-448.
[17] Cao L Y. Practical method for determining the minimum embedding dimension of a scalar time series[J]. Physica D: Nonlinear Phenomena, 1997, 110(1-2): 43-50.
[18] Chen B H, Li J P, Ding R Q. Nonlinear local Lyapunov exponent and atmospheric predictability research[J]. Science in China Series D: Earth Sciences, 2006, 49(10): 1111-1120.
[19] 丁瑞强, 李建平. 误差非线性的增长理论及可预报性研究[J]. 大气科学, 2007, 31(4): 571-576.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】