融合多源求参结果的开采沉陷稳健预计方法  

Steady Mining Subsidence Prediction Method by Fusing Multi-source Parameters Identification Results

作　　者：王磊 魏涛 池深深 陈元非 

机构地区：[1]安徽理工大学测绘学院,安徽淮南232001 [2]中国矿业大学国土环境与灾害监测国家测绘地理信息局重点实验室,江苏徐州221116

出　　处：《煤矿安全》2017年第11期222-225,共4页Safety in Coal Mines

基　　金：国家自然科学基金资助项目(41602357;41474026);安徽省博士后基金资助项目(2014B019);安徽高校自然科学研究资助项目(KJ2016A190)

摘　　要：为了克服单一求参模型由于求参结果不可靠、不精准造成的开采沉陷预计结果不稳健问题,重点研究了融合多源求参模型参数的开采沉陷稳健预计方法,并基于最小二乘原理推导了通用模型,构建了基于融合模矢法、遗传算法和模拟退火求参结果的开采沉陷预计新模型,并对模型的可行性进行了模拟实验验证。实验结果表明:新模型下沉预计误差及水平移动预计误差明显优于基于单一模矢法、遗传算法、模拟退火法反演的参数的下沉预计误差;融合多源求参模型参数的开采沉陷稳健预计方法具有一定抗差能力。In order to overcome the problem of unreliability and inaccuracy results of the single parameter identification model leading to the unsteady of mining subsidence prediction results, steady mining subsidence prediction method by fusing multi-source parameter identification model is studied in this paper, a new model for mining subsidence prediction is established by integrating the parameters identification results of model vector method, genetic algorithm and simulated annealing based on the least square principle, and the feasibility of the model is simulated. The experimental results show that the subsidence error and horizontal displacement error of the new prediction model is better than the estimated error of parameters inversion based on single mode vector method, genetic algorithm, simulated annealing method; new prediction method by fusing multi-source mining subsidence prediction parameters identification model has a certain anti-error ability.

关 键 词：概率积分法 最小二乘 求参模型 融合方法 开采沉陷预计 

分 类 号：TD327[矿业工程—矿井建设]