Page 124 - 2024年第55卷第11期
P. 124
[16] 房昱纬,吴振君,盛谦,等.基于超前钻探测试的隧道地层智能识别方法[J].岩土力学,2020,41(7):
2494 - 2503.
[17] XIAOH H,YANG W K,HU J,etal.Significanceandmethodology:Preprocessingthebigdataformachine
learningonTBM performance[J].UndergroundSpace,2022,7(4):680 - 701.
[18] 余红玲,王晓玲,任炳昱,等.土石坝渗流性态分析的 IAO - XGBoost集成学习模型与预测结果解释 [J].
水利学报,2023,54(10):1195 - 1209.
[19] 刘华吉,孙红林,张占荣,等.基于随钻参数的砂岩与砂质泥岩地层分界面智能识别[J].隧道建设(中英
文),2023,43(S1):304 - 312.
[20] 王小龙,张军.钻孔数据处理与三维显示软件设计及应用[J].煤矿安全,2021,52(10):130 - 135.
[21] LIJB,LIPY,GUOD,etal.Advancedpredictionoftunnelboringmachineperformancebasedonbigdata[J].
GeoscienceFrontiers ,2021,12(1):331 - 338.
[22] 张晓菁,刘攀,周丽婷,等.考虑水文模型参数时变的水库水位多预见期实时预报———以水布垭水库为例
[ J].水利学报,2023,54(4):426 - 438.
[23] ZHANGP,CHENRP,WUHN.Real - timeanalysisandregulationofEPBshieldsteeringusingRandomForest
[J].AutomationinConstruction,2019,106:102860.
[24] 丁佳鑫,郭永峰,米丽娜.高斯噪声和 Lévy噪声激励下欠阻尼周期势系统的相变行为[J].山东大学学报
(理学版),2023,58(8):111 - 117.
[25] 张刚,宋莹,张天骐.Levy噪声驱动下指数型单稳系统的随机共振特性分析[J].电子与信息学报,2017,
39(4):893 - 900.
[26] 胡芳凝,王忠,刘超群.快速滑动平均滤波在 PPG信号去噪中的应用[J].物联网技术,2019,9(6):41 - 43.
[27] 肖浩汉,陈祖煜,徐国鑫,等.基于 GRU算法的盾构掘进参数预测———以成都地铁 19号线为例[J].长
江科学院院报,2023,40(1):123 - 131.
[28] 姜瑶,徐宗学,王静.基于年径流序列的五种趋势检测方法性能对比[J].水利学报,2020,51(7):845 - 857.
[29] ZHANGH,CHENL,QUY,etal.Supportvectorregressionbasedongrid - searchmethodforshort - term wind
powerforecasting [J].JournalofAppliedMathematics,2014.doi:10.1155?2014?835.
[30] 孟品超,李学源,贾洪飞,等.基于滑动平均法的轨道交通短时客流实时预测 [J].吉林大学学报 (工学
版),2018,48(2):448 - 453.
Researchonpreprocessingmethodsformonitoringdrillingdata
XIAOHaohan,CAORuilang,WANGYujie,ZHAOYufei,SUNYanpeng
(ChinaStateKeyLaboratoryofSimulationandRegulationofWaterCycleinRiverBasin,
ChinaInstituteofWaterResourcesandHydropowerResearch,Beijing 100048,China)
Abstract:Toaddresschallengesassociatedwithprocessingmulti - dimensionalheterogeneousmonitoringdrilling
data,thisstudyconductsananalysisofthecharacteristicsofmulti - boreholedrillingdataandintroducesadatapre
processingmethodleveragingstatisticalanalysisandmachinelearningtechniques.Thismethodologybeginswithan
examinationoftheoriginaldrillingdata ’sfeatures,establishescriteriaforextractingstabledrillingstagedata,de
finesparametersforidentifyingabnormaldrillingdata ,andassessestheeffectivenessofvariousmethodsforrepai
ringmissingdrillingdataandreducingnoise.Theresultsindicatethatthetwo - pointlinearinterpolationmethoda
chievedthelowesterrorindexandthemosteffectivemissingdatainterpolation.Additionally ,theButterworthfilter
demonstratedoptimalfilteringperformanceacrossvarioustypesofnoise.Subsequently ,alightweightautomatic
preprocessingsoftwarefordrillingdataisdeveloped ,anditsefficacyinswiftlycompletingtaskssuchasdataextrac
tion,classification,repair,andnoisereductionisvalidatedthroughengineeringcasestudies.Thisendeavorfur
nishesadependabletheoreticalframeworkandempiricaldatasupportforpracticalapplicationsinengineering.
Keywords:digitaldrilling;datapreprocessing;machinelearning;missingdatainterpolation;noisereduction
(责任编辑:李 娜)
— 1 3 0 —
9