摘要
提出一种预测网格传输速率的方法。该方法基于混沌时间序列重构相空间理论,采用加权一阶局域法,对具有自相似性的网格GridFTP传输速率进行预测。仿真结果表明,该方法预测精度高,能够取得良好的预测结果。
In grid, predicting the GridFTP delivery rate between two random grid nodes is very helpful for selecting efficient delivery. This paper put forward a new predicting method for gird delivering rate, which based on chaotic time series phase space reconstruction theories. Adding-weight one-rank local-region was adopted and the GridFTP rate that has self-similarity was predicted. Simulation result showed that this method can get high predicting precision and better predicting result.
出处
《电力科学与工程》
2008年第4期16-19,共4页
Electric Power Science and Engineering
