Collaborative Research: NSWP--Machine Learning and Data Assimilation for Real-Time Radiation Belt Forecasting

合作研究：NSWP--用于实时辐射带预报的机器学习和数据同化

• 批准号: 1131869
• 负责人: Xinlin Li
• 金额: $ 17万
• 依托单位: University of Colorado at Boulder
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2011
• 资助国家: 美国
• 起止时间: 2011-07-15 至 2016-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1131869&HistoricalAwards=false
• 关键词: Collaborative; Research; NSWP; Machine; Learning

This project is a collaborative effort with scientists at the Los Alamos National Laboratory to develop a real-time model for specifying and forecasting the state of Earth's radiation belt. The project will utilize data assimilation methods in conjunction with physical models and modern machine learning techniques to examine the acceleration, transport and loss processes for energetic electrons in the radiation belt. The resulting model will include confidence limits on the system-wide forecasts, making the results useful for operational space weather predicitions. The project will benefit society by making improved radiation-belt predictions at reduced operational costs and the forecasts will be made freely available online.The project includes a strong educational and teaching component through the participation of graduate students at the University of Colorado. In addition the research will be integrated with the Los Alalmos Space Science Outreach (LASSO) program, which offers teacher training to underrepresented groups and individuals with disabilities.

该项目是与洛斯阿拉莫斯国家实验室的科学家合作开发一种实时模型，用于指定和预测地球辐射带的状态。该项目将利用数据同化方法，结合物理模型和现代机器学习技术，研究辐射带中高能电子的加速、传输和损失过程。由此产生的模型将包括全系统预报的置信度，使结果对业务空间天气预报有用。该项目将以更低的运营成本做出更好的辐射带预测，从而使社会受益，这些预测将在网上免费提供。该项目包括通过科罗拉多大学研究生参与的强大的教育和教学部分。此外，这项研究将与LosAlalmos空间科学推广(LASSO)方案相结合，该方案为代表性不足的残疾群体和个人提供教师培训。