基于最优权组合方法的采矿区沉降预测  

Prediction of overall settlement of mining area based on optimal weight combination method

作　　者：徐少富 XU Shaofu(Hunan Corps of China Building Materials Industrial Geological Exploration Center,Zhuzhou Hunan 412000,China)

机构地区：[1]中国建筑材料工业地质勘查中心湖南总队,湖南株洲412000

出　　处：《北京测绘》2023年第10期1425-1430,共6页Beijing Surveying and Mapping

基　　金：国家自然科学基金(42104017)。

摘　　要：针对现有单一模型只能满足1个点预测的要求,缺乏采矿区整体下沉研究与精度评价的问题。本文利用组合预测的思想,以成熟的神经网络模型与非等间隔灰色模型为预测单模型,并以某矿区全倾向观测数据与半走向观测数据为基础,以最优权组合思想实现两种单模型的有效结合。采用概率积分法求参的方法验证工作面沉陷预测精度,采用预测值与实际值求参时所得结果相差指标在合理范围内,验证了最优权组合预测方法在矿区沉降预测中的可行性与适用性。as the existing single model can only meet the requirement of one point prediction,there is a lack of research and accuracy evaluation on the overall subsidence of the mining area.In this paper,combined with the idea of combination prediction,the mature neural network model and non equal interval model are used as the single prediction model,and based on the trend observation data and half strike observation data of a mining area,the optimal weight combination model is used to realize the effective combination of the two single models.The probability integral method is used to verify the accuracy of the combined prediction model in the prediction of the overall settlement of the mining area.The calculated results of the actual value and the predicted value are not much different.The feasibility and applicability of the optimal weight combination prediction method in the mining subsidence prediction are verified.

关 键 词：开采沉降 非等间隔灰色模型 神经网络模型 最优权组合 概率积分法 

分 类 号：P258[天文地球—测绘科学与技术]