逐步回归和BP神经网络模型的枯季月径流预测

枯季径流是工农业用水的重要来源,分析和预报流域枯季来水情况,可为科学制定用水方案、合理调配水资源提供依据。运用逐步回归模型和BP神经网络模型分别对盘龙河流域枯季月径流进行拟合和预报分析,并采用相关系数、相对误差、合格率对两个模型预测精度进行比较。结果表明BP神经网络模型预测精度更高,预测结果精度满足规范要求,更适用于盘龙河流域枯期月径流的预测。

Stepwise Regression and BP Neural Network Model for Monthly Runoff Forecast in Dry Season

Dry season runoff is an important source of water for industry and agriculture.The analysis and forecast of water inflow in dry season can provide basis for scientific water use scheme and rational allocation of water resources.In this paper,stepwise regression model and BP neural network model are used to fit and forecast the dry season monthly runoff of Panlong River Basin,and the forecast accuracy of the two models is compared by correlation coefficient,relative error and pass rate.The results show that the prediction accuracy of BP neural network model is higher and the accuracy of prediction results meets the specification requirements,and it is more suitable for the prediction of monthly runoff in dry season of Panlong River Basin.