| 改进BP算法在风暴潮极值水位预报中的应用 |
| |
| 引用本文： | 纪芳,侯一筠.改进BP算法在风暴潮极值水位预报中的应用[J].中国海洋大学学报(自然科学版),2011,41(5):120-123. |
| |
| 作者姓名： | 纪芳 侯一筠 |
| |
| 作者单位： | 1. 中国科学院海洋研究所,山东青岛266071;中国科学院研究生院,北京100049
2. 中国科学院海洋研究所,山东青岛,266071 |
| |
| 基金项目： | 山东省908专项(SD908-02-08)任务项目资助 |
| |
| 摘 要： | 提出1种基于改进BP算法的风暴潮极值水位预报方法。首先针对现有人工神经网络BP算法存在的缺陷,引入Cauchy训练方法计算连接权的调整,解决BP算法的局部极小点和网络瘫痪的问题。然后结合山东海区水文观测站每年的台风风暴潮资料,应用该改进算法建立台风强度、风速和台风移速与风暴潮极值水位的非线性网络预报模型,对该站极值水位进行预报。实验结果表明,该算法具有很好的稳定性和准确性,为风暴潮预报分析提供了1种可靠的技术手段。
|
| 关 键 词： | 人工神经网络 BP算法 风暴潮 预报模型 |
Application of Improved BP Neural Network Method in Storm Surges Forecast |
| |
| JI Fang,HOU Yi-Jun.Application of Improved BP Neural Network Method in Storm Surges Forecast[J].Periodical of Ocean University of China,2011,41(5):120-123. |
| |
| Authors: | JI Fang HOU Yi-Jun |
| |
| Institution: | JI Fang1,2,HOU Yi-Jun1(1.Institute of Oceanology of Chinese Academy of Sciences,Qingdao 266071,China,2.Graduate University of Chinese Academic of Science Admission Information,Beijing 100049,China) |
| |
| Abstract: | Directed against the shortcomings of local minimum and paralysis of the BP net,this paper presents an improved BP algorithm by combining BP algorithm and Cauchy training.The direct calculation is provided by BP algorithm and Cauchy algorithm provides the random part.According to the storm surge data of hydrological observation stations in a sea area of Shandong each year,we establish the nonlinear prediction model of typhoon intensity,wind speed,and typhoon moving speed and highest water levels of storm sur... |
| |
| Keywords: | artificial neural network BP net storm surge neural network model |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
