预览加载中，请您耐心等待几秒...

基于Matlab和BP神经网络的爆破振动预测系统.docx

预览
1/2
2/2

在线预览结束，喜欢就下载吧，查找使用更方便

下载提示 文本预览

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

基于Matlab和BP神经网络的爆破振动预测系统 1.Introduction Blastingvibrationisanimportantfactorinminingengineering,constructionengineering,andgeologicalengineering.Itsimpactontheenvironmentcannotbeignored.Inordertoreducetheimpactofblastingvibration,itisnecessarytoaccuratelypredicttheblastingvibrationbeforetheblastingprocess.Withthedevelopmentofartificialintelligencetechnology,theapplicationofartificialneuralnetworksinblastingvibrationpredictionhasbecomeahotresearchtopic.Inthispaper,weproposeablastingvibrationpredictionsystembasedonMATLABandBPneuralnetwork. 2.LiteratureReview Therehavebeenmanystudiesonblastingvibrationprediction.Traditionalmethodsincludeempiricalformulas,regressionanalysis,andstatisticalmethods.However,thesemethodshavelimitationsintermsofpredictionaccuracyandgeneralizationability.Withthedevelopmentofneuralnetworktechnology,manyresearchershaveappliedneuralnetworkstothepredictionofblastingvibration,andachievedsatisfactoryresults.Amongthem,theBPneuralnetworkhasbeenwidelyusedduetoitspowerfulnonlinearfittingabilityandstrongrobustness. 3.Methodology Theproposedblastingvibrationpredictionsystemmainlyconsistsoftwomodules:dataprocessingmoduleandpredictionmodule.Inthedataprocessingmodule,datapreprocessingiscarriedouttoeliminateoutliersandfilternoise.Thefiltereddataisthennormalizedtoensuretheconsistencyofinputdata.Inthepredictionmodule,BPneuralnetworkisusedtopredicttheblastingvibration.Theinputlayeroftheneuralnetworkconsistsoftheinputparameterssuchasthechargingweight,thedistancefromtheblastingpointtothemonitoringpoint,andthegeologyandtopographyoftheblastingarea.Thehiddenlayerisusedtoprocesstheinputdata,andtheoutputlayerisusedtooutputthepredictedblastingvibration. 4.ResultsandDiscussion Totesttheeffectivenessoftheproposedsystem,experimentaldatafromblastinginacertainminingareawereusedtotrainandtesttheBPneuralnetwork.Thedatawasrandomlydividedintotwoparts,70%fortrainingand30%fortesting.TheperformanceoftheBPneuralnetworkwasevaluatedusingthemeanabsoluteerror,rootmeansquareerror,andcorr