基于PCA和BP神经网络的径流预测

发布时间：2018-06-26 13:38

本文选题：主成分回归 + 主成分分析　；参考：《中国沙漠》2016年04期

【摘要】：径流预测为流域水资源的合理开发利用与统筹配置提供依据。运用多元线性回归、主成分回归、BP神经网络及主成分分析和BP神经网络相结合的方法,对新疆呼图壁河流域石门水文站2009—2011年各月径流量进行预测,并采用相关系数、确定性系数及均方根误差对各模型预测精度进行比较。结果表明:(1)神经网络等智能算法具有高速寻优的能力,对短时间尺度的月径流量的预测结果较好;(2)主成分回归等常规算法能充分反映出某地区径流的年际的稳定性,对全年径流总量的模拟精度较高;(3)主成分分析和BP神经网络相结合的方法,提高了神经网络的收敛速度,同时降低了局部极值的影响,优于简单的BP神经网络,适用于呼图壁河月径流量预测。
[Abstract]:Runoff prediction provides basis for rational exploitation and overall allocation of water resources in river basin. By means of multivariate linear regression, principal component regression (PCA) and BP neural network (PCA) combined with BP neural network, the monthly runoff of Shimen Hydrologic Station in the Hutubi River Basin of Xinjiang from 2009 to 2011 is predicted, and the correlation coefficient is adopted. Deterministic coefficient and root mean square error are used to compare the prediction accuracy of each model. The results show that: (1) the intelligent algorithm such as neural network has the ability of high speed optimization, and the prediction result of monthly runoff on a short time scale is better; (2) the conventional algorithms such as principal component regression can fully reflect the interannual stability of runoff in a certain area. (3) the method of combining principal component analysis and BP neural network improves the convergence speed of neural network and reduces the influence of local extremum, which is superior to the simple BP neural network. It is suitable for forecasting monthly runoff of Hutubi River.
【作者单位】：新疆大学资源与环境科学学院;新疆大学教育部绿洲生态重点实验室;新疆大学干旱生态环境研究所;新疆大学干旱半干旱区可持续发展国际研究中心;
【基金】：水利部公益性行业科研专项(201301103) 国家自然科学基金项目(41130531,41171023)
【分类号】：P338

【相似文献】