基于BP神经网络的洞庭湖水位变化的预测.pdfVIP

Journal of Water Resources Research 水资源研究, 2012, 1, 222-226 doi:10.4236/jwrr.2012.14031 Published Online August 2012 (/journal/jwrr.html) Prediction of the Water Level Fluctuation of Dongting * Lake Based on BP Neural Network Yujie Yuan1,2, Jie Liang1,2, Lu Huang1,2, Xun Yu1,2, Yeru Peng1,2, Guangming Zeng1,2# 1College of Environmental Science and Engineering, Hunan University, Changsha 2Key Laboratory of Environmental Biology and Pollution Control, Ministry of Education, Hunan University, Changsha Email: yuanyujiehn@163.com, #zgming@ Received: May 4th, 2012; revised: May 18th, 2012; accepted: May 28th, 2012 Abstract: It is of vital significance to study the water level fluctuation of Dongting Lake especially after the operation of the Three Gorges Reservoir (TGR), which may provide useful data and necessary information for the repairing works of the wetland. In this study, the historical time series of outflow discharge of the TGR and water level of Chenglingji were taken as training samples. Based on Levenberg-Marquardt (LM) algorithm, a BP neural network with four layers was established, which well-expressed the unknown but lit- erally existed relationship between outflow discharge of the TGR and the water level of Chenglingji. Then it was applied to the water level prediction in October 2010 of Chenglingji. It is indicated that the trend of ac- tual value and forecast value are in substantial agreement and the maximum and average errors are 3.89% and 0.91%, respectively. It is shown that BP neural network has fairly good simulation accuracy and can be satis- factorily utilized to predict the water level fluctuation of the Dongting Lake. Keywords: Three Gorges Rese