基于BP神经网络的太原市地面沉降预测及分析  

Land Subsidence Forecast and Analysis of Taiyuan City Based on BP Neural Network

作　　者：周艳萍[1] 

机构地区：[1]榆林学院能源工程学院,陕西榆林719000

出　　处：《地质科技情报》2011年第5期115-118,共4页Geological Science and Technology Information

基　　金：教育部科学技术研究重点项目(104038);中国地质调查局地质调查项目(200310400009)

摘　　要：通过对太原市地面沉降资料、气象水文和地质条件以及地下水等大量资料的分析与整理,考虑到太原市各个沉降中心的沉降趋势不尽相同,对太原市吴家堡、西张、万柏林和下元4个沉降中心分别建立BP神经网络模型,并基于训练好的BP神经网络模型,在太原市地下水开采量的规划方案下,预测了在不同降水保证率下2009—2015年地面沉降的趋势,得出太原市地下水开采量的规划对地面沉降的控制是有效的,其中对地面沉降控制最有效的是下元和吴家堡地区,但对于西张地区,沉降量较历年虽然有缓解,可是仍然属下沉趋势,因此有必要对西张地区的地下水开采方案进行进一步调整;同时发现降水有加速地面沉降的负面作用。Based on the analysis of the massive land subsidence,meteorology and hydrology,geological conditions,and groundwater data in Taiyuan,the paper respectively builds BP neural network for four land subsidence centers with reference to their difference of settlement trend.On the basis of training BP neural network model,the paper forecasts the settlement trend from 2009 to 2015 under the groundwater exploitation planning scheme and different levels of precipitation assurance.It is concluded that the groundwater exploitation planning scheme in Taiyuan is effective for controlling land subsidence.It is the most effective for Xiayuan and Wujiabao subsidence centers,but for Xizhang center,which still tends to be sinking though the settlement reduces.Therefore,it is necessary to adjust groundwater exploitation scheme for Xizhang center.Meanwhile,it reveals that the precipitation accelerates the land subsidence.

关 键 词：地面沉降 BP神经网络 地下水开采量 降水量 

分 类 号：P642.27[天文地球—工程地质学]