Multi-scale RBF Prediction Model of Runoff Based on EMD Method

摘要：

Runoff prediction is an important element in the study field of hydrology and water resources. Point to non-linear, chaotic character and with the noise characteristics Run-off signals, we propose a new model based on empirical mode decomposition (EMD) and the RBF neural network (RBF). First, runoff time series will be broken down into a series of different scales intrinsic mode function imf by EMD, Second, the denoise and phase-space reconstruction will be done. The third, we predict each component by RBF. Finally, we reconstruct the final prediction value by each component Simulation results show that the method have a high accuracy hi denoising and prediction of the runoff sequence.

关键词： EMD denoising phase space reconstruction RBF prediction

作者: SuHui Liu Xinxia

作者单位: China Institute of Water Resources and Hydropower Research Hebei University of Engineering Handan Ch Hebei University of EngineeringHandan China 056021

会议类型: 国际会议

会议名称: Third International Conference on Information and Computing(第三届信息与计算科学国际会议 ICIC 2010)

会议地点: 无锡

会议语种:英文

页码: 296-299

在线出版日期: 2010-06-04（万方平台首次上网日期，不代表论文的发表时间）