基于L-M优化神经网络算法的地下水动态模拟与预测  

The Simulation and Prediction of Groundwater Regime based on the Optimized Algorithm of L-M Neural Network

作　　者：常亮[1] 解建仓[1] 王少波[1] 肖志娟[1] 

机构地区：[1]西安理工大学,陕西西安710048

出　　处：《地下水》2005年第5期380-383,共4页Ground water

摘　　要：本文根据地下水与其影响因素之间存在的非线性映射关系,在BP网络模型的基础上,提出一种L evenberg-M arquart优化神经网络算法,并将其应用于某地地下水的动态模拟与预测。通过与BP算法的仿真结果比较分析,发现该算法稳定性好,收敛速度快,预测精度高。The pape is according to the nonlinear relationship between groundwater and its influenced factors, a Levenberg-Marquart optimized algorithm of BP neural network is proposed, and applied for the simulation and prediction of groundwater regime in some area. Through the analysis of simulation results as compared with BP algorithm, this algorithm has the advantages of fine stability, fast convergence speed and high precision.

关 键 词：地下水动态 神经网络 Levenberg-Marquart算法 

分 类 号：P641.74[天文地球—地质矿产勘探]