引用本文:
【打印本页】   【下载PDF全文】   【View/Add Comment】  【Download reader】  【 Close 】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被：浏览 1638次   下载 1563次 本文二维码信息
码上扫一扫！
分享到： 微信 更多
字体:加大+|默认|缩小-
缺失风浪数据补足的神经网络模型
蒋学炼1,2, 李炎保1
1.天津大学建筑工程学院;2.天津城市建设学院 天津市软土特性与工程环境重点实验室
摘要:
基于波浪数据的完备性对于海岸海洋工程设计而言非常关键，详细阐述了风浪观测数据补足神经网络模型的建立方法，构建了两个网络模型，以已有观测资料为样本进行了验证。结果表明，两个网络的训练效果均很好，且单输出目标的分层模拟要优于多输出目标的单层模拟。表明了利用人工神经网络推导缺失波浪条件的可行性。
关键词:  波浪观测  缺失  补足  人工神经网络
DOI：
分类号:
基金项目:国家自然科学基金项目(50779045)
An artificial neural network (ANN) model for supplement of deficient wave observation
JIANG Xue-lian,LI Yan-bao
Abstract:
The complete wave observation is very important for the design of coastal and ocean structures. First, the process of establishing a wave prediction model based on artificial neural network (ANN) technique is described in detail. Then, two models are established and verified by employing historical observation data. Comparisons between the results of two models show that the training results of two networks match the samples well and network with single output is better than network with multiple outputs. Owing to its simplicity and acceptable accuracy, this model can be used as a new tool to obtain the lost wave data in observation.
Key words:  wave observation  deficiency  supplement  artificial neural network (ANN)
Copyright ©  Editorial Office for Marine Sciences Copyright©2008 All Rights Reserved
Supervised by: Chinese Academy of Sciences (CAS)   Sponsored by: Institute of Oceanology, CAS
Address: 7 Nanhai Road, Qingdao, China.  Postcode: 266071  Tel: 0532-82898755  E-mail: bjb@qdio.ac.cn
Technical support: Beijing E-Tiller Co.,Ltd.