【摘要】 为准确可靠预测煤层厚度和构造煤厚度,以淮北矿业股份有限公司芦岭煤矿Ⅱ六采区8煤层作为研究区域,提出一种基于多目标随机森林(MTRF)方法的煤层厚度和构造煤厚度预测模型,并将预测结果与单目标预测(ART)和BP神经网络方法对比。利用粒子群算法结合遗传算法的混合优化算法对随机森林模型和单目标预测模型中的参数进行优化,分别建立参数优化的GAPSO-RF和GAPSO-ART模型。通过实例验证所提方法相比对比方法具有更好的泛化性和稳定性。更多还原

【Abstract】 To accurately and reliably predict the thicknesses of coal seam and tectonic deformed coal,aprediction model of coal seam thickness and tectonic deformed coal thickness based on multi-objective random forest(MTRF)method was proposed.Taking No.8 coal seam of No.6 mining area of Luling Coal Mine of Huaibei Mining Co.,Ltd.as the research area,its results were compared with that of the single target prediction(ART)and BP neural network methods.In the experiment,the particle swarm optimization algorithm and genetic algorithm hybrid optimization algorithm were used to optimize the experimental parameters of random forest and single target prediction methods.The case study shows that the proposed method has better generalization and stability than the comparative methods.更多还原