| 基于支持向量回归机的南黄海浒苔分布面积预测模型 |
| |
| 引用本文： | 韦佳, 何世钧, 周汝雁, 李亚光, 何培民. 基于支持向量回归机的南黄海浒苔分布面积预测模型[J]. 环境工程学报, 2015, 9(6): 3046-3050. doi: 10.12030/j.cjee.20150684 |
| |
| 作者姓名： | 韦佳 何世钧 周汝雁 李亚光 何培民 |
| |
| 作者单位： | 1. 上海海洋大学信息学院, 上海 201306; 2. 上海海洋大学水产与生命学院, 上海 201306; 3. 上海海洋大学海洋科学研究院海洋生态环境与修复研究所, 上海 201306 |
| |
| 基金项目： | 国家科技支撑计划项目课题(2012BAC07B03) 上海海洋大学研究生科研基金资助 |
| |
| 摘 要： | 绿潮作为一种新型的海洋灾害,已经引起了各个国家的重视.依据2012年南黄海海域浒苔遥感监测分布面积数据,选取了温度、天气状况、风向、风力、浪高5种影响浒苔扩散的气候因子,建立了基于SVR的浒苔分布面积预测模型,并与经典的最近邻点插值模型、线性插值模型、3次样条函数插值模型和分段3次Hermite插值模型进行了回归效果的对比.分析结果表明,基于SVR的浒苔分布面积预测模型能够为浒苔遥感数据的插补提供一种方法,且回归效果优于传统的回归方法,为浒苔的防治提供辅助决策信息.
|
| 关 键 词： | 支持向量回归机 浒苔 分布面积预测 经典回归模型 |
| 收稿时间： | 2014-10-04 |
Support vector regression on distribution area forecast of Enteromorpha prolifera in the southern Yellow Sea |
| |
| Wei Jia, He Shijun, Zhou Ruyan, Li Yaguang, He Peimin. Support vector regression on distribution area forecast of Enteromorpha prolifera in the southern Yellow Sea[J]. Chinese Journal of Environmental Engineering, 2015, 9(6): 3046-3050. doi: 10.12030/j.cjee.20150684 |
| |
| Authors: | Wei Jia He Shijun Zhou Ruyan Li Yaguang He Peimin |
| |
| Affiliation: | 1. College of information Technology, Shanghai Ocean University, Shanghai 201306, China; 2. College of Fisheries and Life Science, Shanghai Ocean University, Shanghai 201306, China; 3. Institute of Marine Science, Shanghai Ocean University, Shanghai 201306, China |
| |
| Abstract: | As a novel marine disasters, the green tides has attracted the attention of many countries. We built up a predictive model of Enteromorpha prolifera distribution area based on SVR according to the 2012 Enteromorpha prolifera remote sensing data in the Southern Yellow Sea, picked up temperature, weather, wind direction, wind power and wave height, 5 factors which have impacts on the diffusion of Enteromorpha prolifera and made a regression effect comparison with nearest point interpolation model, linear interpolation model, cubic splines interpolation model and piecewise cubic hermite interpolation model. The analytical result shows that the predictive model of Enteromorpha prolifera distribution area based on SVR can provide a method for the interpolation of Enteromorpha prolifera remote sensing and has a better regression effect than the traditional regression method, and provides assistant decision making information for the Enteromorpha prolifera control. |
| |
| Keywords: | SVR Enteromorpha prolifera distribution area prediction classical regression model |
|
| 点击此处可从《环境工程学报》浏览原始摘要信息 |
|
点击此处可从《环境工程学报》下载全文 |
|
