基于BP神经网络的龙羊峡水库年末消落水位控制研究

Water level control of Longyangxia reservoir based on BP at the end of year

龙羊峡水库长期处于低水位运行,严重影响了其综合效益。运用逐步回归的方法寻找影响水库年末消落水位的主要因素,以此为基础,建立了控制龙羊峡水库年末消落水位的BP神经网络模型,并采用长系列资料对逐步回归模型和BP神经网络模型的预测结果进行了比较和误差分析。结果表明,BP神经网络模型优于逐步回归模型。

The Longyangxia reservoir has been operating at a low water level for a long time,which has affected the benefits of Longyangxia.In this paper,the main factors related to water level at the end of year of Longyangxia reservoir are discussed by using regress method.The BP model is also introduced for controlling Longyangxia reservoir water level.The prediction results are compared and analyzed between regress model and BP model.Results indicate that the BP model is more reasonable and feasible.