论文部分内容阅读
为提高径流预测的准确性,提出一种经验模态分解EMD(Empirical Mode Decomposition)与径向基函数RBF(radial basis function)神经网络的径流预测方法。该方法利用EMD将漳泽水库径流序列分解为5个固有模态函数和1个残余项,然后将各分量作为RBF网络的输入对径流进行预测。结果表明:通过EMD分解,预测效果有明显提高,满足规范要求,计算方法可行。预测结果可为漳泽水库防洪抗旱规划,水资源优化调度提供科学的依据。
In order to improve the accuracy of runoff prediction, a new runoff prediction method based on Empirical Mode Decomposition (EMD) and Radial Basis Function (RBF) neural networks is proposed. This method uses EMD to decompose Zhangze reservoir runoff sequence into five intrinsic mode functions and one residual term, and then predicts the runoff as the input of each RBF network. The results show that the EMD factorization can significantly improve the prediction results and meet the requirements of the code. The calculation method is feasible. The prediction results can provide a scientific basis for flood control and drought relief planning and water resources optimization scheduling in Zhangze Reservoir.