开采影响下建的筑物损害程度的人工神经网络预测模型.pdf

第 23 卷 第 4 期 岩石力学与工程学报 23(4)：583～587 2004 年 2 月 Chinese Journal of Rock Mechanics and Engineering Feb.，2004 开采影响下建筑物损害程度的人工神经网络 预测模型* 郭文兵 1，2 吴财芳 3 邓喀中 1 (1 焦作工学院资源与材料工程系 焦作 454000) (2 中国矿业大学环境与测绘学院 徐州 221008) (3 中国矿业大学资源与地球科学学院 徐州 221008) 摘要 在综合分析开采影响下建筑物损坏程度影响因素的基础上，采用自适应 BP 神经网络技术建立了建筑物采 动损坏程度的预测模型。以大量的建筑物采动损坏实例作为学习训练样本和测试样本，对模型预测结果与实际值 进行了对比分析。结果表明，用人工神经网络方法预测建筑物采动损害程度是可行的。为开采影响下建筑物损坏 程度预测和评价探索出了一种新的方法。 关键词 地下工程，采动损害，建筑物，神经网络，智能预测 分类号 TD 823.83，TP 183 文献标识码 A 文章编号 1000-6915(2004)04-0583-05 ARTIFICIAL NEURAL NETWORK PREDICTION MODEL OF MINING-INDUCED DAMAGE OF BUILDINGS Guo Wenbing1，2，Wu Caifang3，Deng Kazhong1 (1Department of Resource and Material Engineering，Jiaozuo Institute of Technology， Jiaozuo 454000 China) (2College of Environment and Spatial Informatics，China University of Mining and Technology， Xuzhou 221008 China ) (3College of Resource and Earth Science，China University of Mining and Technology， Xuzhou 211008 China) Abstract The main factors affecting the mining-induced damage degree of buildings are comprehensively analyzed. Then the model is established to predict the damage degree of buildings by applying the theory of artificial neural network(ANN). Based on a large amount of cases related to buildings damaged by mining，the predicted results of the model and the measured values are compared and analyzed. The results show that it is feasible to predict the mining-induced damage degree of buildings by ANN technology. Key words underground engineering，mining-induced damage，building，artificial neural networks，intelligent pred