基于BP算法的含沙量预测模型研究

Study on prediction model of sediment concentration based on BP algorithm

为提高水库泥沙含量预测的精度,综合考虑水温、水深、流速等因素对预测精度的影响,构建基于BP算法的含沙量预测模型.首先,选用实测的水温、水深、流速数据作为样本数据进行BP神经网络的训练,通过设置预测误差实现对预测模型的约束,完成了含沙量预测模型的构建.当需要进行含沙量预测时,则将测试数据导入到训练好的预测模型中得到含沙量的预测值。仿真结果表明,此含沙量预测模型获得的含沙量值与实测值之间的误差小,预测精度达到了预期目标.

In order to improve precision of sediment concentration prediction, taking into account water temperature, water depth, velocity and other factors on the prediction accuracy, the sediment concentration prediction model based on BP algorithm was constructed. Firstly, water temperature, water depth and velocity data were chosen as training sample. By setting prediction error as constraint condition, the sediment concentration prediction model was completed. When predicting sediment concentration, the data was input into the trained prediction model to obtain the predicted result. The simulation results obtained that the sediment concentration prediction model reaches the expected goal.