欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对采动区建筑物损害程度的影响因素较多且各因素呈现非线性、多重共线性等特点,借鉴统计学理论并结合工程实际,提出采用不同判别准则的多元判别方法对地采诱发建筑物损害效应进行有效识别。综合考虑地质采矿方面和建筑物本身的因素,选取10个影响砖混结构建筑物采动损害程度的因素作为模型的输入,将砖混结构建筑物的损害等级作为模型的输出,以32个建筑物采动损害的工程实例作为学习样本进行训练;依据不同判别准则及Mahalanobis距离判别分析（MDA）和Fisher线性判别分析（FDA）方法,分别建立建筑物采动损害的MDA模型和FDA模型,并利用该模型对6组待判样本实例进行仿真测试。研究结果表明：MDA方法和FDA方法对学习样本的准确率分别为87.5%和93.8%,对测试样本的准确率分别为83.3%和100%;多元判别分析模型判别性能稳健可靠,FDA法比MDA法判识准确性更高、适用性更强,对现有评价地采诱发建筑物损害方法进行了有效验证和补充。
英文摘要：
Based on the fact that mining-induced damage to building is attributed to various nonlinear and multi-collinear factors, using statistics in engineering practice, a multi-variable discriminant analysis model based on different criteria was proposed to effectively identify the influences of mining underground on buildings. Comprehensively considering the geological mining and building factors, ten large factors effecting buildings damage of brick and concrete structure by mining were selected as the proposed model input variables. The damage level of the brick and concrete structure buildings was taken into account for the proposed model output value. 32 typical cases of buildings and structures damaged by mining were used for training data. Based on different criteria, Mahalanobis distance discriminant analysis （MDA） and Fisher linear discriminant analysis （FDA）, MDA model and FDA model of buildings damage by mining were established, and 6 group cases were sentenced to distinguish samples for simulation test of these proposed models. The results show that the accuracies of the MDA and FDA method of learning samples are 87.5% and 93.8%, respectively, and accuracies of the test samples are 83.3% and 100%, respectively. The multivariate discriminant analysis model is featured with robust and reliable discriminant performance, and FDA exceeds MDA with higher accuracy and stronger applicability, which effectively verifies and supplements the existing methods for evaluating building damage induced by underground mining.
同期刊论文项目
同项目期刊论文
期刊信息
位置：