欢迎来到天天文库
浏览记录
ID:31367400
大小:116.50 KB
页数:9页
时间:2019-01-09
《拱坝变形监测预报的随机森林模型及应用》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库。
1、拱坝变形监测预报的随机森林模型及应用 摘要:大坝变形预报对大坝运行安全评估起着至关重要的作用。传统模型预报精度不够、模拟效果不稳定;若大坝变形数据有异常值时,传统机器算法模型识别和处理异常值的灵活性很小,导致预报结果有偏差。为了解决这些问题,首次将随机森林算法运用到大坝变形监测领域,将大坝测点根据随机森林相似性矩阵分成若干个子集,针对每一个子集建立随机森林预测模型,分区建立预测模型更符合工程实际情况。选取拱坝变形作为研究对象,验证所建模型的适用性。结果表明,根据随机森林的相似性矩阵对大坝各测点的分区情况符合物理和工程实际意义,对各分区子
2、集测点利用随机森林模型建立的预测模型,与支持向量机、BP神经网络模型相比,预测结果精度较高、模型稳定性好,为大坝变形监测提供了新思路。 关键词:拱坝变形;监控模型;监测点分区;随机森林;变形预测 中图分类号:TU196.1文献标志码:A文章编号: 16721683(2016)06011606 Randomforestmodelandapplicationofarchdam′sdeformationmonitoringandprediction LUOHao1,2,GUOShengyong2,BAOWeimin1 1.Colle
3、geofWaterResourcesandHydrology,HohaiUniversity,Nanjing210098,China; 2.YalongRiverHydropowerCompanyLtd,Chengdu9610051,China) Abstract:Damdeformationpredictionplaysanimportantroleinthesafetyassessmentofdamoperation.Traditionalmodelslackforecastingprecisionandthesimulation
4、effectisnotstableenough.Besides,ifabnormalvaluesofdamdeformationexist,traditionalmachinealgorithmmodellackstheflexibilityofdealingwiththeseabnormaldata,whichwillleadtothedeviationoftheforecastingresults.Inordertosolvetheseproblems,randomforestalgorithmwasintroducedtothefi
5、eldofdamdeformationmonitoringforthefirsttime.Similaritymatrixofrandomforestwasappliedtodividedamdeformationmonitoringpointsintoseveralparts.Randomforestspredictionmodelwasestablishedforeachpart,whichwillavoidthedefectsoftraditionalmodelssuchasmodelingofsinglepointorusingt
6、hesamemodelforalldeformationmonitoringpoints.Establishingforecastingmodelfordifferentpartsofdamwasmoreinlinewithengineeringpractice.Deformationdataofonearchdamwasanalyzedandthefeasibilityofrandomforestmodelwasverified.Theresultsshowedthatpartitionofdamdeformationpointsbas
7、edonsimilaritymatrixofrandomforestconformedtothephysicalandengineeringpracticalsignificance.ComparedwithsupportvectormachineandBPneuralnetworkmodel,thepredictionmodelofrandomforestsforeachpart9hadthehigherpredictionprecisionandstability,whichprovidedanewapproachintheareao
8、fdamsafetymonitoring. Keywords:archdamdeformation;monitoringmodel;partitionsofmonitoringpoints;
此文档下载收益归作者所有