西太平洋副高形态指数的分解重构与集成预测

doi:10.11867/j.issn.1001-8166.2004.04.0572

地球科学进展 ›› 2004, Vol. 19 ›› Issue (4): 572 -578. doi: 10.11867/j.issn.1001-8166.2004.04.0572

张韧 ^1，2;董兆俊 ¹;陈奕德 ¹;何金海 ²

1.解放军理工大学气象学院，江苏南京 211101；2.南京气象学院大气科学系，江苏南京 210044

收稿日期:2003-04-09 修回日期:2003-08-29 出版日期:2004-08-01
通讯作者: 张韧（1963-），男，四川峨边人，教授，主要从事热带海洋气象研究. E-mail:E-mail: zren63@hotmail.com
基金资助:
国家自然科学基金项目“西太平洋副高中短期数值预报误差修正研究”（编号：40375019）和“夏季副热带高压变化及其影响天气气候异常的机理”（编号：40135020）资助.

WAVELET DECOMPOSITION AND COMPOSITIVE PREDICTION ON THE MODALITY INDEX OF THE WESTPACIFIC SUBTROPICAL HIGH

ZHANG Ren ^1,2;DONG Zhao-jun ¹;CHEN Yi-de ¹;HE Jin-hai ²

1.Institute of Meteorology,PLA University of Sciences and Technology,Nanjing 211101,China;2.Nanjing Institute of Meteorology, Nanjing 210044, China

Received:2003-04-09 Revised:2003-08-29 Online:2004-08-01 Published:2004-08-01

下载PDF ( 1544 )

用小波分解和自适应神经模糊推理系统(ANFIS)相结合的方法，建立了西太平洋副热带高压形态指数月、季时间尺度的集成预报模型。由于小波分解可在信号的频域—时域内自由伸缩，准确地分解和重构带通、低通信号，因而能将复杂的副高指数时间序列分解为相对简单的周期分量信号，既简化了系统结构，又突出了信号特征。随后基于ANFIS模糊系统的非线性、容错性、自适应性和联想学习功能，建立各分量信号的独立预报模型，最后对分量预报结果进行集成。试验结果表明，该方法在保留预报对象主要特征的前提下，有效降低了预报难度，预报准确率和预报时效均较传统方法有明显的改进和提高。

关键词: 副高指数, 小波分解, 模糊推理, 集成预测

Based on the method of associating wavelet decomposition with adaptive neurofuzzy inference system (ANFIS), a compositive prediction model on the modality index of the westPacific subtropical high(WPSH) on the monthlyseasonal scale was established. Signals can be freely extended/shrinked in frequency time domain and any passband and passlow frequency branch can be accurately produced and reconstructed by means of wavelet decomposition. Therefore, the complex WPSH modality index time series signals can be separated into several relative simple bandpass signals, which both simplify the system structure and stand out the chief characters of signals. Subsequently, the independent prediction model of the decomposed signals were established based on the advantages of ANFIS model, such as nonlinear, bearingerror, selfadapting and associationlearning and the independent predicted results were integrated finally. The test results showed that under the premise of keeping the main characters of forecast body, the prediction difficulty on WPSH system had effectively been decreased, the precision and durative of the compositive prediction model were evidently improved and promoted compared with that of traditional prediction technique. 

Key words: WPSH index, Wavelet decomposition, Fuzzy inference, Compositive prediction.

中图分类号:

P456

［1］Feng Minxue(冯民学),Chen Biyun(陈必云), Yuan Ruiqin(袁瑞琴),et al.West Pacific subtropical high forecast operation system based on artificial neural network［J］. Scientia Meteorologica [LL]Sinica(气象科学),1998,18(4):396-401(in Chinese).
［2］Zhang Ren(张韧). Diagnosis and prediction research on the Pacific subtropical high based on the forward-propagation network approximation［J］.Acta Scientia Atmospherica Sinica（大气科学）,2001,25(5):650-660(in Chinese).
［3］Zhang Ren(张韧),Jiang Guorong(蒋国荣),Yu Zhihao(余志豪),et al.Establish prediction model of the Pacific subtropical high by ANN method［J］.Quarterly Journal of Applied Meteorologica（应用气象学报）,2000,11(4):474-482(in Chinese).
［4］Zhang Ren,Yu Zhihao.Neural network BP model approximation and prediction of complicated weather systems［J］.Acta Meteorologica Sinica,2001,15(1):105-115.
［5］Jin Long(金龙),Chen Ning(陈宁),Lin Zhenshan(林振山),et al. Research and comparison of ANN-based consensus forecasting［J］. Acta Meteorologica Sinica（气象学报）,1999,57(2):198-207(in Chinese).
［6］Tang Xunchang(唐洵昌),Ge Wenzhong(葛文忠)，Wang Xiangdong(王向东), et al.ANN techniques for weather radar identified severe convective weather［J］.Scientia Meteorologica Sinica（气象科学）, 1997，17(4)：393-399(in Chinese).
［7］Hornik K, Stinchcombe M, White H. Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks［J］.Neural Networks,1990,3(4):551-560.
［8］Mallat S. A theory for multiresolution signal decomposition: The wavelet representation［J］.IEEE Transactions on Pattern Analysis and Mathchine Intelligence,1989,11(7):674-693.
［9］Li Chongyin(李崇银).Introduction of Climate Dynamics［M］.Beijing:China Meteorology Press,1995(in Chinese)
［10］Li Shiyong(李士勇).Fuzzy Control, Neurocontrol and Intelligent Cybernetics［M］.Haerbin: Haerbin Industry University Press,1998(in Chinese). 
［11］Roger Jang J S, Sun C T, Mizutani E. Neuro-Fuzzy and Soft Computing: A Computational Approach to Learning and Machine Intelligence［M］.Englewood Cliffs:Prentice-Hall,Inc,1997.
［12］Takagi T, Sugeon M. Fuzzy identification and its application to modeling and control［J］.IEEE SMC,1985,15(1):116-132.

[1]	张韧，王继光，蒋国荣，周树道. 非线性模糊识别及其在海温异常检测中的应用[J]. 地球科学进展, 2002, 17(4): 470-476.

阅读次数

全文

摘要