欢迎访问《金属矿山》杂志官方网站,今天是 分享到:

金属矿山 ›› 2009, Vol. 39 ›› Issue (12): 93-97.

• 机电与自动化 • 上一篇    下一篇

基于GIS与ANN的金川二矿地表移动预测

邓清海1,马凤山2,袁仁茂3,张丽萍1   

  1. 1.山东科技大学;2.中国科学院地质与地球物理研究所;3.中国地震局地质研究所
  • 出版日期:2009-12-15 发布日期:2011-04-24
  • 基金资助:

    中国科学院知识创新工程重要方向性项目(编号:KZCX2-YW-113),国家自然科学基金项目(编号:40702048),山东科技大学科学研究“春蕾计划”项目(编号:2008AZZ016)。

Ground Movement Prediction of No.2 Nickel Mine Area in Jinchuan Based on GIS and ANN

Deng Qinghai1,Ma Fengshan2,Yuan Renmao3,Zhang Liping1   

  1. 1.Shandong University of Science and Technology;2.Institute of Geology and Geophysics, CAS;3.Institute of Geology, China Earthquake Administration
  • Online:2009-12-15 Published:2011-04-24

摘要: 利用金川二矿区多年的GPS监测数据,探讨了GIS与人工神经网络技术相结合进行地表移动定量预测的方法和思路。借助Avenue语言,实现了在GIS平台下进行地表移动神经网络预测的整个过程,包括样本设计、网络设计与训练、网络测试与网络预测等;其中,数据处理、测试结果对比和预测结果分析由GIS软件完成,采用VB调用Matlab6.5实现Elman神经网络预测模型,并通过Avenue编程将其集成到GIS系统中。研究结果表明,利用GIS支持下的神经网络模型对地表移动进行预测,具有理论上的可行性和现实意义,说明GIS和人工神经网络技术在开采沉陷预计领域中具有广阔的应用前景。

关键词: 地表移动, 预测, 地理信息系统, 人工神经网络

Abstract: Based on GPS monitoring data on No.2 Nickel Mine Area in Jinchuan, a new idea of quantitative prediction of ground surface movement by means of GIS and ANN(Artificial Neural Network) was put forward and the specific method was given.The whole process of ground surface movement prediction based on ANN may be carried out in GIS platform by Avenue programming, including designing data mode, designing and training network framework, testing network framework, forecasting by network, and so on.Such works as data processing, contrasting testing results and analysing forecasting results were achieved by GIS.The Elman ANN prediction model was realized by the technique of Matlab 6.5 program with VB6.0, which was integrated in GIS by Avenue programming.The results stated that the ANN prediction model supported by GIS had a theoretical feasibility and realistic significance in predicting ground surface movement, and GIS and ANN had a wide application prospect in prediction of exploitation subsidence.

Key words: Ground surface movement, Prediction, GIS, ANN(Artificial Neural Network)