A Way To Predict The Solar Wind Velocity Base On SVR Method

摘要：

We apply a new neural network technique, support vector regression (SVR), to forecast the solar wind (SW) velocity. In our work, we choose the SW data during the nine years from 1998 to 2006, which includes the periods of the SW speed variation associated with high speed streams from coronal hole and coronal mass ejections (CMEs). We fred that the new approach is accurate and reliable in forecasting SW velocity. Based on the error difference, it can be concluded that the SVR technique can lend it itself to future space weather forecasting models.

关键词： SVR method Solar Wind space weather forecast

作者: Cong Huang Dan-Dan Liu

作者单位: National Center For Space Weather, CMA, Beijing, 100081, China

会议类型: 国际会议

会议名称: 2010 International Conference on Application of Mathematics and Physics(2010国际数理科学与气象学术研讨会暨2010空间天气学研讨会 AMP2010)

会议地点: 南京

会议语种:英文

页码: 25-29

在线出版日期: 2010-05-08（万方平台首次上网日期，不代表论文的发表时间）