基于BP神经网络的灰色自记忆径流预测模型

提高径流预测精度在于对有限径流序列所包含的信息进行充分挖掘.针对灰色、灰色自记忆建模中存在的缺陷,基于组合预测思想,借助BP神经网络方法处理灰色自记忆模型仍存在的误差,建立了基于BP神经网络的灰色自记忆径漉预测模型,并应用于新疆和田玉龙喀什河同古孜洛克水文站年径流预测中,结果表明改进后的灰色自记忆模型具有更好的拟合与预报精度,是一种有效的征流预测方法.

Grey self-memory model based on BP neural network for annual runoff prediction

The key problem of improving the accuracy of runoff prediction is safficiently to dig the information included in the sample series.For the defect of the grey and grey self-memory model,on account of integrative prediction,the BP neural network is used to deal with the error existed in grey self-memory model,then the grey self-memory model based on BP neural network is developed.It is shown that the model has better prediction accuracy and may be used for annual runoff prediction.